fix(ocr-pipeline): increase LLM timeout to 300s and disable qwen3 thinking

- Add /no_think tag to prompt (qwen3 thinking mode causes massive slowdown)
- Increase httpx timeout from 120s to 300s for large vocab tables
- Improve error logging with traceback and exception type

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

klausur-service/backend/cv_vocab_pipeline.py

@@ -4354,11 +4354,13 @@ Antworte NUR mit dem korrigierten JSON-Array. Kein erklaerener Text.
 Fuer jeden Eintrag den du aenderst, setze "corrected": true.
 Fuer unveraenderte Eintraege setze "corrected": false.
 
+/no_think
+
 Eingabe:
 {_json.dumps(table_lines, ensure_ascii=False, indent=2)}"""
 
     t0 = time.time()
-    async with httpx.AsyncClient(timeout=120.0) as client:
+    async with httpx.AsyncClient(timeout=300.0) as client:
         resp = await client.post(
             f"{_OLLAMA_URL}/api/chat",
             json={

klausur-service/backend/ocr_pipeline_api.py

@@ -1420,8 +1420,9 @@ async def run_llm_review(session_id: str, request: Request):
     try:
         result = await llm_review_entries(entries, model=model)
     except Exception as e:
-        logger.error(f"LLM review failed for session {session_id}: {e}")
-        raise HTTPException(status_code=502, detail=f"LLM review failed: {e}")
+        import traceback
+        logger.error(f"LLM review failed for session {session_id}: {type(e).__name__}: {e}\n{traceback.format_exc()}")
+        raise HTTPException(status_code=502, detail=f"LLM review failed ({type(e).__name__}): {e}")
 
     # Store result inside word_result as a sub-key
     word_result["llm_review"] = {