feat: Add Document Crawler & Auto-Onboarding service (Phase 1.4)

New standalone Python/FastAPI service for automatic compliance document
scanning, LLM-based classification, IPFS archival, and gap analysis.
Includes extractors (PDF, DOCX, XLSX, PPTX), keyword fallback classifier,
compliance matrix, and full REST API on port 8098.

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

document-crawler/tests/test_classifier.py

@@ -0,0 +1,36 @@
+"""Tests for keyword fallback classifier."""
+
+import pytest
+from classifiers.keyword_fallback import keyword_classify
+
+
+def test_vvt_detection():
+    text = "Verzeichnis von Verarbeitungstaetigkeiten gemaess Art. 30 DSGVO"
+    result = keyword_classify(text, "vvt.pdf")
+    assert result["classification"] == "VVT"
+    assert result["confidence"] <= 0.3
+
+
+def test_tom_detection():
+    text = "Technisch-organisatorische Massnahmen: Zutrittskontrolle, Zugangskontrolle, Verschluesselungskonzept"
+    result = keyword_classify(text, "toms.docx")
+    assert result["classification"] == "TOM"
+
+
+def test_dse_detection():
+    text = "Datenschutzerklaerung: Informationspflichten nach Art. 13 DSGVO"
+    result = keyword_classify(text, "datenschutz.pdf")
+    assert result["classification"] == "DSE"
+
+
+def test_unknown_document():
+    text = "Lorem ipsum dolor sit amet"
+    result = keyword_classify(text, "random.pdf")
+    assert result["classification"] == "Sonstiges"
+    assert result["confidence"] == 0.1
+
+
+def test_confidence_capped():
+    text = "Verarbeitungsverzeichnis Art. 30 Kategorie betroffener Personen Datenkategorien Zweck der Verarbeitung"
+    result = keyword_classify(text, "vvt_complete.pdf")
+    assert result["confidence"] <= 0.3