refactor: move document domain core to lib/document/

Moves ~25 components, utils (search, filename, groupDocuments,
documentStatusLabel, validateFile), bulkSelection store, and
TranscriptionSection sub-component. Fixes broken relative imports.

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

frontend/src/lib/document/filename.ts

@@ -0,0 +1,87 @@
+import { isoToGerman } from '$lib/utils/date';
+
+export interface FilenameParseResult {
+	/** ISO format: YYYY-MM-DD */
+	dateIso?: string;
+	/** "Firstname Lastname" — order reversed from filename convention */
+	personName?: string;
+	/** Ready-to-use title, e.g. "Hans Mueller (12.03.1965)" */
+	suggestedTitle?: string;
+}
+
+// A date token is either YYYY-MM-DD or YYYYMMDD with a plausible month/day range.
+function tryParseDate(s: string): string | undefined {
+	if (/^\d{4}-\d{2}-\d{2}$/.test(s)) {
+		const m = parseInt(s.slice(5, 7));
+		const d = parseInt(s.slice(8, 10));
+		if (m >= 1 && m <= 12 && d >= 1 && d <= 31) return s;
+	} else if (/^\d{8}$/.test(s)) {
+		const m = parseInt(s.slice(4, 6));
+		const d = parseInt(s.slice(6, 8));
+		if (m >= 1 && m <= 12 && d >= 1 && d <= 31)
+			return `${s.slice(0, 4)}-${s.slice(4, 6)}-${s.slice(6, 8)}`;
+	}
+	return undefined;
+}
+
+const NAME_PART = /^\p{L}+$/u;
+
+/**
+ * Parses a structured filename and extracts a date and person name.
+ *
+ * Supported conventions (date-first or date-last, compound last names supported):
+ *   YYYY-MM-DD_Lastname_Firstname.ext
+ *   YYYYMMDD_Lastname_Firstname.ext
+ *   YYYYMMDD_de_Gruyter_Walter.ext   ← compound last name: lastName="de Gruyter"
+ *   Lastname_Firstname_YYYY-MM-DD.ext
+ *   Lastname_Firstname_YYYYMMDD.ext
+ *   de_Gruyter_Walter_YYYYMMDD.ext   ← compound last name: lastName="de Gruyter"
+ *
+ * Algorithm: split on "_", identify the date token (first or last segment),
+ * treat the outermost remaining segment as firstName, rest as lastName parts.
+ * Returns {} for anything that doesn't match cleanly.
+ */
+export function parseFilename(filename: string): FilenameParseResult {
+	const dot = filename.lastIndexOf('.');
+	if (dot < 0) return {}; // no extension — not a real file
+	const stem = filename.slice(0, dot);
+	const parts = stem.split('_');
+
+	// Minimum: date + at least one lastName segment + firstName = 3 parts
+	if (parts.length < 3) return {};
+
+	let dateIso: string;
+	let nameParts: string[];
+
+	const dateFromFirst = tryParseDate(parts[0]);
+	if (dateFromFirst) {
+		dateIso = dateFromFirst;
+		nameParts = parts.slice(1);
+	} else {
+		const dateFromLast = tryParseDate(parts[parts.length - 1]);
+		if (!dateFromLast) return {};
+		dateIso = dateFromLast;
+		nameParts = parts.slice(0, -1);
+	}
+
+	// Need at least lastName + firstName after removing the date
+	if (nameParts.length < 2) return {};
+
+	// All name segments must be pure letters (covers umlauts via \p{L})
+	if (!nameParts.every((p) => NAME_PART.test(p))) return {};
+
+	const firstName = nameParts[nameParts.length - 1];
+	const lastName = nameParts.slice(0, -1).join(' ');
+	const personName = `${firstName} ${lastName}`;
+	const suggestedTitle = `${personName} (${isoToGerman(dateIso)})`;
+
+	return { dateIso, personName, suggestedTitle };
+}
+
+export function stripExtension(filename: string): string {
+	return filename.replace(/\.[^/.]+$/, '');
+}
+
+export function bulkTitleFromFilename(filename: string): string {
+	return stripExtension(filename).replace(/[_-]+/g, ' ').trim();
+}