fix(nlp-service): eliminate false-positive person matches from dirty DB records

- Wire _EXTRA_SPAN_STOPS into _extract_persons_and_role so German function
  words (im, seine, ihre, dem, …) terminate name spans — fixes "Clara im"
  and "seine Kinder" leaking into personNames
- Add _NON_NAME_TOKENS filter in PersonMatcher.load() to skip DB records
  whose first_name contains prepositions or possessives — filters 290 bad
  records (annotations like "an seine Eltern", "Eltern in", place references
  like "Enkel Cram aus Mexiko") that were causing exact Pass-2 matches
- Remove spaCy model downloads from Dockerfile (no longer needed after the
  DB-backed matcher rewrite)

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

nlp-service/Dockerfile

@@ -9,11 +9,6 @@ RUN apt-get update && apt-get install -y --no-install-recommends \
 COPY requirements.txt .
 RUN pip install --no-cache-dir -r requirements.txt
 
-# Bake models into the image — no volume needed, ~350 MB total
-RUN python -m spacy download de_core_news_sm \
- && python -m spacy download en_core_web_sm \
- && python -m spacy download es_core_news_sm
-
 COPY . .
 
 RUN useradd --no-create-home --shell /usr/sbin/nologin --uid 1001 nlp \

nlp-service/extractor.py

@@ -66,6 +66,22 @@ _DATE_BETWEEN: dict[str, frozenset[str]] = {
     "es": frozenset({"entre"}),
 }
 
+# ── Extra span-termination tokens (function words that cannot be in a name) ──
+
+_EXTRA_SPAN_STOPS: dict[str, frozenset[str]] = {
+    # German articles, possessives, and particles that end a name span
+    "de": frozenset({
+        "im", "am", "beim", "zum", "zur",
+        "dem", "den", "des",
+        "sein", "seine", "seinen", "seiner",
+        "ihr", "ihre", "ihrem", "ihren", "ihrer",
+        "unser", "unsere", "unseren",
+        "über", "auch", "oder", "und",
+    }),
+    "en": frozenset(),
+    "es": frozenset({"el", "la", "los", "las", "su", "sus", "mi"}),
+}
+
 # ── Stopword lists ────────────────────────────────────────────────────────────
 
 _STOPWORDS: dict[str, frozenset[str]] = {
@@ -138,7 +154,7 @@ def _extract_persons_and_role(
         return [], "any"
 
     preps = _ALL_PERSON_PREPS[lang]
-    stops = preps | _DATE_BEFORE[lang] | _DATE_AFTER[lang] | _DATE_BETWEEN[lang]
+    stops = preps | _DATE_BEFORE[lang] | _DATE_AFTER[lang] | _DATE_BETWEEN[lang] | _EXTRA_SPAN_STOPS[lang]
     matches = m.find_in_query(query, preps, stop_tokens=stops)
 
     person_names = [text for text, _ in matches]

nlp-service/person_matcher.py

@@ -8,6 +8,22 @@ from rapidfuzz import fuzz, process
 _PUNCT_RE = re.compile(r"[^\w\s\-]", re.UNICODE)
 _YEAR_PAT = re.compile(r"^\d{4}$")
 
+# Tokens that cannot appear in a real person's first name — used to filter DB
+# records that are annotations or descriptions masquerading as persons.
+_NON_NAME_TOKENS: frozenset[str] = frozenset({
+    # German prepositions
+    "an", "in", "im", "am", "aus", "von", "vom", "nach", "zu", "zum", "zur",
+    "für", "bei", "beim", "mit", "über", "unter", "durch", "gegen", "ohne",
+    "bis", "seit", "des", "dem", "den",
+    # German possessives / pronouns
+    "sein", "seine", "seinen", "seiner",
+    "ihr", "ihre", "ihren", "ihrem",
+    # English prepositions
+    "for", "from", "by", "of",
+    # Spanish prepositions
+    "del", "por", "para",
+})
+
 
 class PersonMatcher:
     """Match person name fragments from free-text queries against known persons.
@@ -30,6 +46,10 @@ class PersonMatcher:
         for first, last in rows:
             first = (first or "").strip()
             last = (last or "").strip()
+            # Skip records whose first_name contains function words — these are
+            # annotations or descriptions in the DB, not real person names.
+            if any(w in _NON_NAME_TOKENS for w in first.lower().split()):
+                continue
             for variant in _name_variants(first, last):
                 key = variant.lower()
                 if key not in seen: