M25: expose provenance-aware knowledge statistics

backend/app/services/knowledge/ragcore.py

@@ -1,9 +1,12 @@
 from __future__ import annotations
 
+from pathlib import Path
+
 import httpx
 
 from app.core.config import get_settings
 from app.services.knowledge import EvidenceState, GroundedAnswer, KnowledgeHealth, SourceCard
+from app.services.knowledge.procedures import iter_procedure_documents
 
 _GROUNDED_ANSWERABILITY = {"answerable", "partially_answerable"}
 
@@ -140,6 +143,10 @@ class RAGcoreKnowledgeProvider:
         except (httpx.HTTPError, ValueError) as exc:
             available = False
             detail = f"RAGcore unavailable: {type(exc).__name__}: {exc}"
+        source_document_count = sum(
+            document.language == language
+            for document in iter_procedure_documents(Path(self._settings.knowledge_dir))
+        )
         return KnowledgeHealth(
             provider=self.name,
             available=available,
@@ -150,6 +157,11 @@ class RAGcoreKnowledgeProvider:
             # RAGcore's retrieval API has no corpus-size endpoint. Unknown is explicit
             # so the UI never turns this into the misleading claim "0 procedures".
             document_count=None,
+            source_document_count=source_document_count,
+            reported_synced_document_count=None,
+            reported_failed_document_count=None,
+            last_sync_at=None,
+            statistics_state="not_reported",
         )
 
     def ask(self, question: str, correlation_id: str, language: str = "en-GB") -> GroundedAnswer: