fix(profiler): fix B2G false positive + add consulting/manufacturing

- Remove generic B2G keywords (behörde, amt, öffentlich) that match in
  every DSI due to "Aufsichtsbehörde", "Amtsgericht", "veröffentlichen"
- Remove "server" from it_services (too generic, appears in every DSI)
- Add consulting, manufacturing, media industries
- Add B2B fallback for GmbH/AG without B2C signals
- Add 10 ground truth files for unified compliance check

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

backend-compliance/compliance/services/business_profiler.py

@@ -40,14 +40,17 @@ _B2C_KEYWORDS = [
 
 _B2B_KEYWORDS = [
     "unternehmen", "geschaeftskunden", "geschäftskunden", "gewerblich",
-    "auftrag", "auftraggeber", "auftragnehmer", "geschaeftspartner",
-    "geschäftspartner", "firmenkunde", "b2b",
+    "auftraggeber", "auftragnehmer", "geschaeftspartner",
+    "geschäftspartner", "firmenkunde", "b2b", "industriekunden",
+    "beratung", "consulting", "dienstleistung", "engineering",
 ]
 
 _B2G_KEYWORDS = [
-    "behoerde", "behörde", "koerperschaft", "körperschaft", "oeffentlich",
-    "öffentlich", "gemeinde", "amt", "stadtverwaltung", "landesbehoerde",
-    "landesbehörde", "kommunal",
+    "koerperschaft des oeffentlichen rechts", "körperschaft des öffentlichen rechts",
+    "gemeinde", "stadtverwaltung", "landesbehoerde", "landesbehörde",
+    "kommunal", "buergerservice", "bürgerservice", "rathaus",
+    "landesamt", "bundesamt", "oeffentliche verwaltung", "öffentliche verwaltung",
+    "oeffentlicher dienst", "öffentlicher dienst",
 ]
 
 _NONPROFIT_KEYWORDS = [
@@ -93,14 +96,17 @@ _EDITORIAL_KEYWORDS = [
 ]
 
 _INDUSTRY_KEYWORDS = {
-    "it_services": ["software", "saas", "cloud", "hosting", "server", "api", "app"],
+    "it_services": ["software", "saas", "cloud", "hosting", "api", "plattform"],
     "retail": ["shop", "warenkorb", "versand", "lieferung", "einzelhandel"],
     "healthcare": ["arzt", "praxis", "patient", "gesundheit", "therapie", "klinik"],
     "legal": ["kanzlei", "rechtsanwalt", "mandant", "anwalt"],
     "craft": ["handwerk", "meister", "werkstatt", "montage", "gewerk"],
-    "public": ["behoerde", "behörde", "kommune", "verwaltung", "buerger", "bürger"],
+    "public": ["kommune", "stadtverwaltung", "buergerservice", "bürgerservice", "rathaus"],
     "finance": ["bank", "versicherung", "finanz", "kredit", "anlage"],
     "education": ["schule", "bildung", "unterricht", "lehrplan", "schueler", "schüler"],
+    "consulting": ["beratung", "consulting", "schulung", "seminar", "gutachten", "audit"],
+    "manufacturing": ["fertigung", "produktion", "maschinenbau", "anlagenbau", "zulieferer"],
+    "media": ["redaktion", "verlag", "medien", "journalismus", "presse"],
 }
 
 _TRACKING_SERVICES = {
@@ -197,8 +203,16 @@ async def detect_business_profile(documents: dict[str, str]) -> BusinessProfile:
         total = sum(max(0, v) for v in scores.values())
         profile.confidence = round(best_val / total, 2) if total > 0 else 0.5
     else:
-        profile.business_type = "unknown"
-        profile.confidence = 0.2
+        # Fallback: GmbH/AG without B2C signals → assume B2B
+        has_company = any(kw in full_text for kw in [
+            "gmbh", "ag ", "ohg", "kg ", "ug ", "gbr",
+        ])
+        if has_company and b2c_score <= 0:
+            profile.business_type = "b2b"
+            profile.confidence = 0.4
+        else:
+            profile.business_type = "unknown"
+            profile.confidence = 0.2
 
     # ── ODR (Online-Streitbeilegung) ─────────────────────────────
     # Required for B2C with online shop (EU Regulation 524/2013)