fix: make _build_context and execute_chat async for hybrid search

- Changed _build_context to async def
- Changed execute_chat to async def in RagChatUseCase
- Changed execute_chat to async def in AgentRouterUseCase
- Updated all callers to use await
- Fixes SyntaxError: 'await' outside async function

src/api/routes/rag.py

@@ -61,7 +61,7 @@ def direct_search(
         raise HTTPException(status_code=500, detail=str(e))
 
 @router.post("/chat/test", response_model=ChatResponse)
-def chat_test(
+async def chat_test(
     request: ChatRequest,
     agent_router_use_case: AgentRouterUseCase = Depends(get_agent_router_use_case)
 ):
@@ -70,7 +70,7 @@ def chat_test(
     # Get endpoint from environment variable or use default
     os.getenv("RAG_ENDPOINT", "/rag/chat/test")
     
-    result = agent_router_use_case.execute_chat(request)
+    result = await agent_router_use_case.execute_chat(request)
     return result
 
 @router.post("/chat/stream")

src/core/use_cases/agent_router_use_case.py

@@ -23,7 +23,7 @@ class AgentRouterUseCase:
         print(f"DEBUG: Intent Classification: {intent} for query: '{query[:80]}'")
         return intent
 
-    def execute_chat(self, request: ChatRequest) -> Dict[str, Any]:
+    async def execute_chat(self, request: ChatRequest) -> Dict[str, Any]:
         intent = self._classify_intent(request.query)
         session_id = request.session_id or "anonymous"
         
@@ -38,7 +38,7 @@ class AgentRouterUseCase:
             }
         else:
             print(f"DEBUG: Routing to NEWS (RAG Pipeline)")
-            return self.rag_chat.execute_chat(request)
+            return await self.rag_chat.execute_chat(request)
 
     async def execute_stream(self, request: ChatRequest, is_guest: bool = False, user_id: int = None) -> AsyncGenerator[str, None]:
         intent = self._classify_intent(request.query)

src/core/use_cases/rag_chat_use_case.py

@@ -316,7 +316,7 @@ JSON:"""
             print(f"DEBUG: [{lang_code}] search failed: {e}")
             return []
 
-    def _build_context(self, query: str, top_k: int, source_filter=None, language_filter=None, days_back=None) -> Tuple[str, List[Dict[str, Any]]]:
+    async def _build_context(self, query: str, top_k: int, source_filter=None, language_filter=None, days_back=None) -> Tuple[str, List[Dict[str, Any]]]:
         # ── Step 1: Single LLM call — intent extraction + multilingual translation ──
         expanded_query = query
 
@@ -609,7 +609,7 @@ JSON:"""
             "intent": f"intent:{query_prefix_hash}"
         }
 
-    def execute_chat(self, request: ChatRequest) -> Dict[str, Any]:
+    async def execute_chat(self, request: ChatRequest) -> Dict[str, Any]:
         print(f"DEBUG: execute_chat called with query: {request.query}")
         session_id = request.session_id or "default_session"
         
@@ -634,7 +634,7 @@ JSON:"""
         print("DEBUG: Cache MISS - Layer 1 (full response) - proceeding with RAG pipeline")
         history_text = self._get_history_text(session_id)
 
-        context_text, final_sources = self._build_context(
+        context_text, final_sources = await self._build_context(
             request.query, 
             request.top_k,
             request.source_filter,
@@ -730,7 +730,7 @@ Answer:"""
     async def execute_stream(self, request: ChatRequest, is_guest: bool = False, user_id: int = None) -> AsyncGenerator[str, None]:
         session_id = request.session_id or "default_session"
         history_text = "" if is_guest else self._get_history_text(session_id)
-        context_text, final_sources = self._build_context(
+        context_text, final_sources = await self._build_context(
             request.query, request.top_k, request.source_filter, request.language_filter, getattr(request, 'days_back', None)
         )