Improve cancel generation responsiveness

- Add additional cancel checks before each yield in the streaming loop
- Ensures generation stops immediately when cancel is triggered, preventing partial responses from being displayed
- Complements the StoppingCriteria to provide more robust cancellation

app.py

@@ -545,6 +545,8 @@ def chat_response(user_msg, chat_history, system_prompt,
                     history.append({'role': 'assistant', 'content': answer_buf})
                 else:
                     history[-1]['content'] = thought_buf
+                if cancel_event.is_set():
+                    break
                 yield history, debug
                 continue
 
@@ -560,6 +562,8 @@ def chat_response(user_msg, chat_history, system_prompt,
                     history.append({'role': 'assistant', 'content': answer_buf})
                 else:
                     history[-1]['content'] = thought_buf
+                if cancel_event.is_set():
+                    break
                 yield history, debug
                 continue
 
@@ -568,6 +572,8 @@ def chat_response(user_msg, chat_history, system_prompt,
                 history.append({'role': 'assistant', 'content': ''})
             answer_buf += text
             history[-1]['content'] = answer_buf
+            if cancel_event.is_set():
+                break
             yield history, debug
 
         gen_thread.join()