refactor(search): remove NLP/smart-search feature entirely (#772)

## Summary

- Removes the NLP/smart-search feature completely — the feature was too unreliable and slow; users get better results with the regular search filters
- Deletes the entire backend `search/` package (NlSearchController, NlQueryParserService, NlpClient, NlSearchRateLimiter — 14 classes + 6 test classes)
- Deletes the `nlp-service/` Python microservice (FastAPI, rapidfuzz, DB-backed person matching)
- Removes all frontend NL search components: SmartModeToggle, SmartSearchStatus, InterpretationChipRow, DisambiguationPicker, chip-types, theme-chip-removal
- Strips smart-mode logic from SearchFilterBar and documents/+page.svelte
- Removes `SMART_SEARCH_UNAVAILABLE` / `SMART_SEARCH_RATE_LIMITED` error codes from backend, frontend types, and all three i18n files (de/en/es)
- Removes `nlp-service` container and `APP_NLP_BASE_URL` from both docker-compose files
- Removes Ollama/NLP Prometheus scrape job and Grafana dashboard
- Deletes ADRs 028 (×2), 034, 035

## Test plan

- [ ] Backend compiles: `cd backend && ./mvnw compile -q` → BUILD SUCCESS
- [ ] Frontend server tests pass: `cd frontend && npm run test -- --project=server`
- [ ] No NLP/smart-search references remain in source: `grep -r "SmartSearch\|NlSearch\|nlp-service\|SMART_SEARCH" backend/src frontend/src`
- [ ] `docker compose config` validates both compose files
- [ ] Search page loads, filter bar works, no smart-mode toggle visible

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-authored-by: Marcel <marcel@familienarchiv>
Reviewed-on: #772

backend/src/main/java/org/raddatz/familienarchiv/exception/ErrorCode.java

@@ -135,12 +135,6 @@ public enum ErrorCode {
     /** The merge target is a descendant of the source tag. 400 */
     TAG_MERGE_INVALID_TARGET,
 
-    // --- NL Search ---
-    /** Ollama is unreachable or timed out. 503 */
-    SMART_SEARCH_UNAVAILABLE,
-    /** NL search rate limit exceeded (5 requests per user per minute). 429 */
-    SMART_SEARCH_RATE_LIMITED,
-
     // --- Generic ---
     /** Request validation failed (missing or malformed fields). 400 */
     VALIDATION_ERROR,

backend/src/main/java/org/raddatz/familienarchiv/search/NlQueryInterpretation.java

@@ -1,26 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import io.swagger.v3.oas.annotations.media.Schema;
-
-import java.time.LocalDate;
-import java.util.List;
-
-public record NlQueryInterpretation(
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        List<PersonHint> resolvedPersons,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        List<PersonHint> ambiguousPersons,
-        LocalDate dateFrom,
-        LocalDate dateTo,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        List<String> keywords,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        List<TagHint> resolvedTags,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        String rawQuery,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        boolean keywordsApplied,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        boolean tagsApplied
-) {
-}

backend/src/main/java/org/raddatz/familienarchiv/search/NlQueryParserService.java

@@ -1,216 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import lombok.RequiredArgsConstructor;
-import lombok.extern.slf4j.Slf4j;
-import org.raddatz.familienarchiv.document.DocumentSearchResult;
-import org.raddatz.familienarchiv.document.DocumentService;
-import org.raddatz.familienarchiv.document.DocumentSort;
-import org.raddatz.familienarchiv.document.SearchFilters;
-import org.raddatz.familienarchiv.exception.DomainException;
-import org.raddatz.familienarchiv.exception.ErrorCode;
-import org.raddatz.familienarchiv.person.NameMatches;
-import org.raddatz.familienarchiv.person.Person;
-import org.raddatz.familienarchiv.person.PersonService;
-import org.raddatz.familienarchiv.tag.Tag;
-import org.raddatz.familienarchiv.tag.TagOperator;
-import org.raddatz.familienarchiv.tag.TagService;
-import org.springframework.data.domain.Pageable;
-import org.springframework.stereotype.Service;
-
-import java.time.LocalDate;
-import java.util.ArrayList;
-import java.util.LinkedHashSet;
-import java.util.List;
-import java.util.UUID;
-
-@Service
-@RequiredArgsConstructor
-@Slf4j
-public class NlQueryParserService {
-
-    private static final int MIN_QUERY = 3;
-    private static final int MAX_QUERY = 500;
-    private static final int MAX_NAME_LENGTH = 200;
-    private static final int MIN_TAG_TERM = 3;
-    private static final int MAX_RESOLVED_TAGS = 10;
-
-    private final OllamaClient ollamaClient;
-    private final PersonService personService;
-    private final DocumentService documentService;
-    private final TagService tagService;
-
-    public NlSearchResponse search(String query, Pageable pageable) {
-        if (query == null || query.length() < MIN_QUERY || query.length() > MAX_QUERY) {
-            throw DomainException.badRequest(ErrorCode.VALIDATION_ERROR,
-                    "Query must be between " + MIN_QUERY + " and " + MAX_QUERY + " characters");
-        }
-
-        OllamaExtraction ext = ollamaClient.parse(query);
-
-        List<String> personNames = ext.personNames() != null ? ext.personNames() : List.of();
-        List<String> keywords = ext.keywords() != null ? ext.keywords() : List.of();
-
-        TagResolution tagResolution = resolveTags(keywords);
-        List<TagHint> resolvedTagHints = tagResolution.hints();
-        List<String> resolvedTagNames = tagResolution.tagNames();
-        List<String> remainingKeywords = tagResolution.remaining();
-
-        NameResolution resolution = resolveNames(personNames);
-
-        if (!resolution.ambiguous().isEmpty()) {
-            NlQueryInterpretation interpretation = new NlQueryInterpretation(
-                    List.of(), resolution.ambiguous(),
-                    ext.dateFrom(), ext.dateTo(),
-                    keywords, List.of(), ext.rawQuery(), false, false);
-            return new NlSearchResponse(DocumentSearchResult.of(List.of()), interpretation);
-        }
-
-        List<PersonHint> resolved = resolution.resolved();
-        List<String> noMatchFragments = resolution.noMatchFragments();
-        List<String> extraFragments = resolution.extraFragments();
-
-        boolean hadStructuredMatch = !resolvedTagHints.isEmpty() || !resolved.isEmpty();
-        String text = buildText(remainingKeywords, noMatchFragments, extraFragments, ext.rawQuery(), hadStructuredMatch);
-
-        if (resolved.size() == 1 && isAnyRole(ext.personRole())) {
-            UUID personId = resolved.get(0).id();
-            DocumentSearchResult docs = documentService.searchDocumentsByPersonId(
-                    personId, ext.dateFrom(), ext.dateTo(), pageable);
-            NlQueryInterpretation interpretation = new NlQueryInterpretation(
-                    resolved, List.of(), ext.dateFrom(), ext.dateTo(), keywords, resolvedTagHints, ext.rawQuery(), false, false);
-            return new NlSearchResponse(docs, interpretation);
-        }
-
-        UUID sender = buildSender(resolved, ext.personRole());
-        UUID receiver = buildReceiver(resolved, ext.personRole());
-
-        boolean tagsApplied = !resolvedTagHints.isEmpty();
-        TagOperator tagOperator = tagsApplied ? TagOperator.OR : TagOperator.AND;
-
-        SearchFilters filters = new SearchFilters(
-                text.isBlank() ? null : text,
-                ext.dateFrom(), ext.dateTo(),
-                sender, receiver,
-                resolvedTagNames, null,
-                null, tagOperator, false);
-
-        DocumentSearchResult docs = documentService.searchDocuments(filters, DocumentSort.DATE, "desc", pageable);
-        boolean keywordsApplied = !text.isBlank();
-        NlQueryInterpretation interpretation = new NlQueryInterpretation(
-                resolved, List.of(), ext.dateFrom(), ext.dateTo(), keywords, resolvedTagHints, ext.rawQuery(), keywordsApplied, tagsApplied);
-        return new NlSearchResponse(docs, interpretation);
-    }
-
-    private NameResolution resolveNames(List<String> personNames) {
-        List<PersonHint> resolved = new ArrayList<>();
-        List<PersonHint> ambiguous = new ArrayList<>();
-        List<String> noMatchFragments = new ArrayList<>();
-        List<String> extraFragments = new ArrayList<>();
-
-        int resolvedIndex = 0;
-        for (String name : personNames) {
-            if (name == null || name.length() > MAX_NAME_LENGTH) {
-                log.debug("Skipping name fragment (too long or null): length={}", name == null ? 0 : name.length());
-                continue;
-            }
-            NameMatches matches = personService.resolveByName(name);
-            List<Person> direct = matches.direct();
-            List<Person> partial = matches.partial();
-
-            if (direct.size() == 1) {
-                Person p = direct.get(0);
-                resolvedIndex++;
-                if (resolvedIndex <= 2) {
-                    resolved.add(new PersonHint(p.getId(), p.getDisplayName()));
-                } else {
-                    extraFragments.add(name);
-                }
-            } else if (direct.size() >= 2) {
-                direct.forEach(p -> ambiguous.add(new PersonHint(p.getId(), p.getDisplayName())));
-            } else if (!partial.isEmpty()) {
-                partial.forEach(p -> ambiguous.add(new PersonHint(p.getId(), p.getDisplayName())));
-            } else {
-                noMatchFragments.add(name);
-            }
-        }
-
-        return new NameResolution(resolved, ambiguous, noMatchFragments, extraFragments);
-    }
-
-    private TagResolution resolveTags(List<String> keywords) {
-        LinkedHashSet<Tag> seen = new LinkedHashSet<>();
-        List<String> remaining = new ArrayList<>();
-
-        for (String kw : keywords) {
-            if (kw == null || kw.length() < MIN_TAG_TERM) {
-                remaining.add(kw);
-                continue;
-            }
-            List<Tag> matches = tagService.findByNameContaining(kw);
-            if (matches.isEmpty()) {
-                remaining.add(kw);
-            } else {
-                seen.addAll(matches);
-            }
-        }
-
-        if (seen.size() > MAX_RESOLVED_TAGS) {
-            log.debug("Keyword matched {} tags; capping at {}", seen.size(), MAX_RESOLVED_TAGS);
-        }
-        List<Tag> capped = seen.size() > MAX_RESOLVED_TAGS
-                ? new ArrayList<>(seen).subList(0, MAX_RESOLVED_TAGS)
-                : new ArrayList<>(seen);
-
-        // safe: entities are detached here; mutation is for DTO projection only, no dirty-check fires
-        tagService.resolveEffectiveColors(capped);
-
-        List<TagHint> hints = capped.stream()
-                .map(t -> new TagHint(t.getId(), t.getName(), t.getColor()))
-                .toList();
-        List<String> tagNames = capped.stream().map(Tag::getName).toList();
-
-        return new TagResolution(hints, tagNames, remaining);
-    }
-
-    private String buildText(List<String> keywords, List<String> noMatchFragments,
-                             List<String> extraFragments, String rawQuery, boolean hadStructuredMatch) {
-        List<String> parts = new ArrayList<>();
-        parts.addAll(keywords);
-        parts.addAll(noMatchFragments);
-        parts.addAll(extraFragments);
-        String text = String.join(" ", parts).strip();
-        if (text.isBlank() && !hadStructuredMatch && rawQuery != null && !rawQuery.isBlank()) {
-            return rawQuery;
-        }
-        return text;
-    }
-
-    private boolean isAnyRole(String role) {
-        return role == null || "any".equals(role) || (!"sender".equals(role) && !"receiver".equals(role));
-    }
-
-    private UUID buildSender(List<PersonHint> resolved, String role) {
-        if (resolved.size() >= 2) return resolved.get(0).id();
-        if (resolved.size() == 1 && "sender".equals(role)) return resolved.get(0).id();
-        return null;
-    }
-
-    private UUID buildReceiver(List<PersonHint> resolved, String role) {
-        if (resolved.size() >= 2) return resolved.get(1).id();
-        if (resolved.size() == 1 && "receiver".equals(role)) return resolved.get(0).id();
-        return null;
-    }
-
-    private record NameResolution(
-            List<PersonHint> resolved,
-            List<PersonHint> ambiguous,
-            List<String> noMatchFragments,
-            List<String> extraFragments
-    ) {}
-
-    private record TagResolution(
-            List<TagHint> hints,
-            List<String> tagNames,
-            List<String> remaining
-    ) {}
-}

backend/src/main/java/org/raddatz/familienarchiv/search/NlSearchController.java

@@ -1,28 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import jakarta.validation.Valid;
-import lombok.RequiredArgsConstructor;
-import org.raddatz.familienarchiv.security.Permission;
-import org.raddatz.familienarchiv.security.RequirePermission;
-import org.springframework.data.domain.Pageable;
-import org.springframework.security.core.annotation.AuthenticationPrincipal;
-import org.springframework.security.core.userdetails.UserDetails;
-import org.springframework.web.bind.annotation.*;
-
-@RestController
-@RequestMapping("/api/search/nl")
-@RequiredArgsConstructor
-public class NlSearchController {
-
-    private final NlQueryParserService nlQueryParserService;
-    private final NlSearchRateLimiter rateLimiter;
-
-    @PostMapping
-    @RequirePermission(Permission.READ_ALL)
-    public NlSearchResponse search(@Valid @RequestBody NlSearchRequest request,
-                                   Pageable pageable,
-                                   @AuthenticationPrincipal UserDetails principal) {
-        rateLimiter.checkAndConsume(principal.getUsername());
-        return nlQueryParserService.search(request.query(), pageable);
-    }
-}

backend/src/main/java/org/raddatz/familienarchiv/search/NlSearchRateLimitProperties.java

@@ -1,12 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import lombok.Data;
-import org.springframework.boot.context.properties.ConfigurationProperties;
-import org.springframework.stereotype.Component;
-
-@Component
-@ConfigurationProperties("app.nl-search.rate-limit")
-@Data
-public class NlSearchRateLimitProperties {
-    private int maxRequestsPerMinute = 5;
-}

backend/src/main/java/org/raddatz/familienarchiv/search/NlSearchRateLimiter.java

@@ -1,46 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import com.github.benmanes.caffeine.cache.Caffeine;
-import com.github.benmanes.caffeine.cache.LoadingCache;
-import io.github.bucket4j.Bandwidth;
-import io.github.bucket4j.Bucket;
-import org.raddatz.familienarchiv.exception.DomainException;
-import org.raddatz.familienarchiv.exception.ErrorCode;
-import org.springframework.stereotype.Service;
-
-import java.time.Duration;
-import java.util.concurrent.TimeUnit;
-
-@Service
-public class NlSearchRateLimiter {
-
-    private final LoadingCache<String, Bucket> byUser;
-    private final int maxRequestsPerMinute;
-
-    public NlSearchRateLimiter(NlSearchRateLimitProperties props) {
-        this.maxRequestsPerMinute = props.getMaxRequestsPerMinute();
-        this.byUser = Caffeine.newBuilder()
-                .expireAfterAccess(1, TimeUnit.MINUTES)
-                .build(key -> newBucket(maxRequestsPerMinute));
-    }
-
-    public void checkAndConsume(String userKey) {
-        if (!byUser.get(userKey).tryConsume(1)) {
-            throw DomainException.tooManyRequests(ErrorCode.SMART_SEARCH_RATE_LIMITED,
-                    "NL search rate limit exceeded for user: " + userKey, 60L);
-        }
-    }
-
-    void resetForTest() {
-        byUser.invalidateAll();
-    }
-
-    private static Bucket newBucket(int limit) {
-        return Bucket.builder()
-                .addLimit(Bandwidth.builder()
-                        .capacity(limit)
-                        .refillGreedy(limit, Duration.ofMinutes(1))
-                        .build())
-                .build();
-    }
-}

backend/src/main/java/org/raddatz/familienarchiv/search/NlSearchRequest.java

@@ -1,11 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import jakarta.validation.constraints.NotBlank;
-import jakarta.validation.constraints.Size;
-
-public record NlSearchRequest(
-        @NotBlank
-        @Size(min = 3, max = 500)
-        String query
-) {
-}

backend/src/main/java/org/raddatz/familienarchiv/search/NlSearchResponse.java

@@ -1,12 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import io.swagger.v3.oas.annotations.media.Schema;
-import org.raddatz.familienarchiv.document.DocumentSearchResult;
-
-public record NlSearchResponse(
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        DocumentSearchResult result,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        NlQueryInterpretation interpretation
-) {
-}

backend/src/main/java/org/raddatz/familienarchiv/search/OllamaClient.java

@@ -1,5 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-public interface OllamaClient {
-    OllamaExtraction parse(String query);
-}

backend/src/main/java/org/raddatz/familienarchiv/search/OllamaExtraction.java

@@ -1,18 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import java.time.LocalDate;
-import java.util.List;
-
-/**
- * Raw structured output from Ollama after parsing and sanitising.
- * personRole is always one of "sender", "receiver", "any" — defensive parsing ensures this.
- */
-record OllamaExtraction(
-        List<String> personNames,
-        String personRole,
-        LocalDate dateFrom,
-        LocalDate dateTo,
-        List<String> keywords,
-        String rawQuery
-) {
-}

backend/src/main/java/org/raddatz/familienarchiv/search/OllamaHealthClient.java

@@ -1,5 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-public interface OllamaHealthClient {
-    boolean isHealthy();
-}

backend/src/main/java/org/raddatz/familienarchiv/search/OllamaProperties.java

@@ -1,15 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import lombok.Data;
-import org.springframework.boot.context.properties.ConfigurationProperties;
-import org.springframework.stereotype.Component;
-
-@Component
-@ConfigurationProperties("app.ollama")
-@Data
-public class OllamaProperties {
-    private String baseUrl;
-    private String model;
-    private int timeoutSeconds = 30;
-    private int healthCheckTimeoutSeconds = 2;
-}

backend/src/main/java/org/raddatz/familienarchiv/search/PersonHint.java

@@ -1,13 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import io.swagger.v3.oas.annotations.media.Schema;
-
-import java.util.UUID;
-
-public record PersonHint(
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        UUID id,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        String displayName
-) {
-}

backend/src/main/java/org/raddatz/familienarchiv/search/RestClientOllamaClient.java

@@ -1,184 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import com.fasterxml.jackson.annotation.JsonIgnoreProperties;
-import com.fasterxml.jackson.annotation.JsonProperty;
-import com.fasterxml.jackson.databind.ObjectMapper;
-import lombok.extern.slf4j.Slf4j;
-import org.raddatz.familienarchiv.exception.DomainException;
-import org.raddatz.familienarchiv.exception.ErrorCode;
-import org.springframework.http.client.JdkClientHttpRequestFactory;
-import org.springframework.stereotype.Service;
-import org.springframework.web.client.RestClient;
-import org.springframework.web.client.RestClientException;
-
-import java.net.http.HttpClient;
-import java.time.Duration;
-import java.time.LocalDate;
-import java.time.Year;
-import java.time.format.DateTimeFormatter;
-import java.time.format.DateTimeParseException;
-import java.util.List;
-import java.util.Map;
-import java.util.Set;
-
-@Service
-@Slf4j
-public class RestClientOllamaClient implements OllamaClient, OllamaHealthClient {
-
-    private static final ObjectMapper MAPPER = new ObjectMapper();
-    private static final Set<String> VALID_ROLES = Set.of("sender", "receiver", "any");
-    private static final int MAX_NAME_LENGTH = 200;
-    private static final int MAX_KEYWORD_LENGTH = 100;
-
-    private static final Map<String, Object> JSON_SCHEMA = Map.of(
-            "type", "object",
-            "required", List.of("personNames", "personRole", "keywords"),
-            "properties", Map.of(
-                    "personNames", Map.of("type", "array", "items", Map.of("type", "string", "maxLength", MAX_NAME_LENGTH)),
-                    "personRole", Map.of("type", "string", "enum", List.of("sender", "receiver", "any")),
-                    "dateFrom", Map.of("type", List.of("string", "null"), "maxLength", 20),
-                    "dateTo", Map.of("type", List.of("string", "null"), "maxLength", 20),
-                    "keywords", Map.of("type", "array", "items", Map.of("type", "string", "maxLength", MAX_KEYWORD_LENGTH))
-            )
-    );
-
-    private final RestClient inferenceClient;
-    private final RestClient healthClient;
-    private final OllamaProperties props;
-
-    public RestClientOllamaClient(OllamaProperties props) {
-        this.props = props;
-
-        HttpClient inferenceHttp = HttpClient.newBuilder()
-                .version(HttpClient.Version.HTTP_1_1)
-                .connectTimeout(Duration.ofSeconds(10))
-                .build();
-        JdkClientHttpRequestFactory inferenceFactory = new JdkClientHttpRequestFactory(inferenceHttp);
-        inferenceFactory.setReadTimeout(Duration.ofSeconds(props.getTimeoutSeconds()));
-        this.inferenceClient = RestClient.builder()
-                .baseUrl(props.getBaseUrl())
-                .requestFactory(inferenceFactory)
-                .build();
-
-        HttpClient healthHttp = HttpClient.newBuilder()
-                .version(HttpClient.Version.HTTP_1_1)
-                .connectTimeout(Duration.ofSeconds(props.getHealthCheckTimeoutSeconds()))
-                .build();
-        JdkClientHttpRequestFactory healthFactory = new JdkClientHttpRequestFactory(healthHttp);
-        healthFactory.setReadTimeout(Duration.ofSeconds(props.getHealthCheckTimeoutSeconds()));
-        this.healthClient = RestClient.builder()
-                .baseUrl(props.getBaseUrl())
-                .requestFactory(healthFactory)
-                .build();
-    }
-
-    @Override
-    public OllamaExtraction parse(String query) {
-        try {
-            OllamaGenerateRequest request = new OllamaGenerateRequest(
-                    props.getModel(), query, JSON_SCHEMA, false);
-            String responseBody = inferenceClient.post()
-                    .uri("/api/generate")
-                    .contentType(org.springframework.http.MediaType.APPLICATION_JSON)
-                    .body(request)
-                    .retrieve()
-                    .body(String.class);
-            return parseOllamaResponse(responseBody, query);
-        } catch (DomainException e) {
-            throw e;
-        } catch (Exception e) {
-            log.warn("Ollama inference failed: {}", e.getClass().getSimpleName());
-            throw DomainException.serviceUnavailable(ErrorCode.SMART_SEARCH_UNAVAILABLE,
-                    "Ollama unavailable: " + e.getClass().getSimpleName());
-        }
-    }
-
-    @Override
-    public boolean isHealthy() {
-        try {
-            healthClient.get().uri("/api/tags").retrieve().toBodilessEntity();
-            return true;
-        } catch (Exception e) {
-            return false;
-        }
-    }
-
-    private OllamaExtraction parseOllamaResponse(String responseBody, String rawQuery) {
-        try {
-            OllamaGenerateResponse response = MAPPER.readValue(responseBody, OllamaGenerateResponse.class);
-            String inner = response.response();
-            if (inner == null || inner.isBlank()) {
-                return fallbackExtraction(rawQuery);
-            }
-            RawOllamaOutput raw = MAPPER.readValue(inner, RawOllamaOutput.class);
-            return toExtraction(raw, rawQuery);
-        } catch (Exception e) {
-            log.warn("Failed to parse Ollama response: {}", e.getClass().getSimpleName());
-            throw DomainException.serviceUnavailable(ErrorCode.SMART_SEARCH_UNAVAILABLE,
-                    "Failed to parse Ollama response: " + e.getClass().getSimpleName());
-        }
-    }
-
-    private OllamaExtraction toExtraction(RawOllamaOutput raw, String rawQuery) {
-        List<String> names = raw.personNames() == null ? List.of() : raw.personNames().stream()
-                .filter(n -> n != null && n.length() <= MAX_NAME_LENGTH)
-                .toList();
-        List<String> keywords = raw.keywords() == null ? List.of() : raw.keywords().stream()
-                .filter(k -> k != null && k.length() <= MAX_KEYWORD_LENGTH)
-                .toList();
-        String role = sanitiseRole(raw.personRole());
-        LocalDate dateFrom = parseDate(raw.dateFrom(), true);
-        LocalDate dateTo = parseDate(raw.dateTo(), false);
-        return new OllamaExtraction(names, role, dateFrom, dateTo, keywords, rawQuery);
-    }
-
-    private OllamaExtraction fallbackExtraction(String rawQuery) {
-        return new OllamaExtraction(List.of(), "any", null, null, List.of(), rawQuery);
-    }
-
-    private String sanitiseRole(String role) {
-        if (role != null && VALID_ROLES.contains(role)) {
-            return role;
-        }
-        log.warn("Unexpected personRole from Ollama: {}", role);
-        return "any";
-    }
-
-    private LocalDate parseDate(String raw, boolean isFrom) {
-        if (raw == null || raw.isBlank()) return null;
-        try {
-            return LocalDate.parse(raw, DateTimeFormatter.ISO_LOCAL_DATE);
-        } catch (DateTimeParseException ignored) {
-        }
-        try {
-            int year = Integer.parseInt(raw.strip());
-            if (year > 1000 && year < 3000) {
-                return isFrom ? Year.of(year).atDay(1) : Year.of(year).atMonth(12).atEndOfMonth();
-            }
-        } catch (NumberFormatException ignored) {
-        }
-        return null;
-    }
-
-    @JsonIgnoreProperties(ignoreUnknown = true)
-    private record OllamaGenerateResponse(String response) {
-    }
-
-    @JsonIgnoreProperties(ignoreUnknown = true)
-    private record RawOllamaOutput(
-            @JsonProperty("personNames") List<String> personNames,
-            @JsonProperty("personRole") String personRole,
-            @JsonProperty("dateFrom") String dateFrom,
-            @JsonProperty("dateTo") String dateTo,
-            @JsonProperty("keywords") List<String> keywords
-    ) {
-    }
-
-    private record OllamaGenerateRequest(
-            String model,
-            String prompt,
-            Object format,
-            boolean stream
-    ) {
-    }
-}

backend/src/main/java/org/raddatz/familienarchiv/search/TagHint.java

@@ -1,14 +0,0 @@
-package org.raddatz.familienarchiv.search;
-
-import io.swagger.v3.oas.annotations.media.Schema;
-
-import java.util.UUID;
-
-public record TagHint(
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        UUID id,
-        @Schema(requiredMode = Schema.RequiredMode.REQUIRED)
-        String name,
-        String color
-) {
-}

backend/src/main/resources/application-dev.yaml

@@ -12,6 +12,3 @@ springdoc:
     enabled: true
     path: /swagger-ui.html
 
-app:
-  ollama:
-    base-url: http://localhost:11434

backend/src/main/resources/application.yaml

@@ -130,18 +130,6 @@ app:
     # The loader maps columns by header name — no positional indices (see ADR-025).
     dir: ${IMPORT_DIR:/import}
 
-  ollama:
-    base-url: http://ollama:11434
-    model: qwen2.5:7b-instruct-q4_K_M
-    # CPU inference: ~18s warm. Higher ceiling absorbs the cold model load on the
-    # first query after an Ollama (re)start before OLLAMA_KEEP_ALIVE pins it.
-    timeout-seconds: 60
-    health-check-timeout-seconds: 2
-
-  nl-search:
-    rate-limit:
-      max-requests-per-minute: 5
-
 ocr:
   sender-model:
     activation-threshold: 100