End a trace for an LLM or chat model run.
Note:
This is the end callback for both run types. Chat models start with
on_chat_model_start, but there is no on_chat_model_end;
completion is routed here for callback API compatibility.
The response.
The run ID.
Additional arguments.
