feat: 7 Analyse-Module auf 100% — Backend-Endpoints, DB-Model, Frontend-Persistenz

Alle 7 Analyse-Module (Requirements → Report) von ~80% auf 100% gebracht:
- Modul 1 (Requirements): POST/DELETE Endpoints + Frontend-Anbindung + Rollback
- Modul 2 (Controls): Evidence-Linking UI mit Validity-Badge
- Modul 3 (Evidence): Pagination (Frontend + Backend)
- Modul 4 (Risk Matrix): Mitigation-UI, Residual Risk, Status-Workflow
- Modul 5 (AI Act): AISystemDB Model, 6 CRUD-Endpoints, Backend-Persistenz
- Modul 6 (Audit Checklist): PDF-Download + Session-History
- Modul 7 (Audit Report): Detail-Seite mit Checklist Sign-Off + Navigation

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

backend-compliance/compliance/api/evidence_routes.py

@@ -46,9 +46,11 @@ async def list_evidence(
     control_id: Optional[str] = None,
     evidence_type: Optional[str] = None,
     status: Optional[str] = None,
+    page: Optional[int] = Query(None, ge=1, description="Page number (1-based)"),
+    limit: Optional[int] = Query(None, ge=1, le=500, description="Items per page"),
     db: Session = Depends(get_db),
 ):
-    """List evidence with optional filters."""
+    """List evidence with optional filters and pagination."""
     repo = EvidenceRepository(db)
 
     if control_id:
@@ -71,6 +73,13 @@ async def list_evidence(
         except ValueError:
             pass
 
+    total = len(evidence)
+
+    # Apply pagination if requested
+    if page is not None and limit is not None:
+        offset = (page - 1) * limit
+        evidence = evidence[offset:offset + limit]
+
     results = [
         EvidenceResponse(
             id=e.id,
@@ -95,7 +104,7 @@ async def list_evidence(
         for e in evidence
     ]
 
-    return EvidenceListResponse(evidence=results, total=len(results))
+    return EvidenceListResponse(evidence=results, total=total)
 
 
 @router.post("/evidence", response_model=EvidenceResponse)