Enhance intent classification and debugging in AI chat

- Added debug print statements in `extract_and_update_intent` to log missing intent fields and the classification check, improving traceability during intent processing.
- Updated the return structure of `extract_and_update_intent` to include the classification result, enhancing the information provided for subsequent processing.
- These changes aim to refine the intent extraction logic and provide better insights into the classification process, ultimately improving user interactions.

ai_chat.py

@@ -268,9 +268,11 @@ async def extract_and_update_intent(state):
     print(f"DEBUG - Final intent state: {state['intent']}")
 
     missing = [f for f in intent_fields if state["intent"].get(f) is None]
+    print(f"DEBUG - Missing intent fields: {missing}")
     if missing:
         # Check if user is asking for properties (using AI classification)
         classification = state.get("classification")
+        print(f"DEBUG - Intent update classification check: '{classification}'")
         if classification == "search_listings":
             # User is asking for properties, don't interrupt with preference questions
             print("DEBUG - User asking for properties, skipping preference questions")
@@ -308,7 +310,7 @@ Message: {user_message}
     classification = resp.content.strip()
     state["classification"] = classification
     print(f"DEBUG - Classification result: '{classification}' for message: '{user_message}'")
-    return {"response": None}
+    return {"classification": classification, "response": None}
 
 async def extract_and_search_properties(state):
     """