Python: Auto Function Invocation Filter is not handled properly for OpenAI models

[moonbox3]

When using the Auto Function Invocation Filter, with a context.terminate condition, the original completions are returned to the caller instead of the auto invocation function result(s).

Item 1:

Non-streaming:

Streaming:

If I run the auto invoke function filter example, I should see "Stop trying to ask me to do math, I don't like it!" as the result, but instead I see a manual tool call as the result (the underlying function was already run, so the result should be returned instead of a tool call):

With the updates shown in the image, I get the correction invocation context back with the injected filter result value of "Stop asking me to do math, I don't like it!"

Item 2:

If using the auto invoke kernel function filter, the function result was never added to the chat history -- thus, in the chat history, there would be a tool call entry, and never a FunctionResult content entry following it up.

We should still add the function result to the chat history so it's persisted, even if terminating with the invocation context function result or the completions result. Need to make sure that the chat history is still included in the response's metadata, similar to how it's done for completions.

[moonbox3]

The chat completions can be simplified to:

if any(result.terminate for result in results if result is not None):
    return chat_history.messages[-1]

since we're adding the function results to the chat every time.