fix: Section classifier strips leading numbers + recognizes German headings

- "5. Soziale Medien" now stripped to "soziale medien" before classification
- Added "soziale medien/netzwerke" as social_media heading pattern
- Fixes etogruppe.com where Social Media section wasn't detected

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

backend-compliance/compliance/api/agent_doc_check_routes.py

@@ -327,7 +327,7 @@ SECTION_TYPE_MAP = [
     (r"^(?:agb|allgemeine geschäftsbedingungen|nutzungsbedingungen)$", "agb"),
     # DSFA MUST be checked BEFORE social_media (both can contain "Social Media")
     (r"datenschutzfolge|dsfa|risikoanalyse", "dsfa"),
-    (r"^social\s*media$", "social_media"),  # Standalone heading "Social Media" = DSE
+    (r"^social\s*media$|^soziale\s+(?:medien|netzwerke)$", "social_media"),
     (r"datenschutzerkl(?:ae|ä)rung.*social|datenschutz\s+f(?:ue|ü)r\s+social", "social_media"),
 ]
 
@@ -415,6 +415,8 @@ def _classify_section(heading: str) -> str | None:
     """Classify a section heading into a document type."""
     import re as _re
     heading_lower = heading.lower().strip()
+    # Strip leading numbers/bullets: "5. Soziale Medien" → "soziale medien"
+    heading_lower = _re.sub(r"^[\d\.\)\-]+\s*", "", heading_lower).strip()
     # Skip known sub-sections
     if heading_lower in SKIP_HEADINGS:
         return None