[openapi server] log exception in exception handler(2/N) (#36201)

Signed-off-by: Andy Xie <andy.xning@gmail.com>

vllm/entrypoints/openai/chat_completion/api_router.py

@@ -50,10 +50,7 @@ async def create_chat_completion(request: ChatCompletionRequest, raw_request: Re
     )
     handler = chat(raw_request)
     if handler is None:
-        base_server = raw_request.app.state.openai_serving_tokenization
-        return base_server.create_error_response(
-            message="The model does not support Chat Completions API"
-        )
+        raise NotImplementedError("The model does not support Chat Completions API")
 
     generator = await handler.create_chat_completion(request, raw_request)