Fix BaseOpenAIChatCompletionClient token usage

[gziz]

Why are these changes needed?

To correctly track the token usage in BaseOpenAIChatCompletionClient, more info on the issue.

Related issue number

#4769

Checks

• I've included any doc changes needed for https://microsoft.github.io/autogen/. See https://microsoft.github.io/autogen/docs/Contribute#documentation to build and test documentation locally.
• I've added tests (if relevant) corresponding to the changes introduced in this PR.
• I've made sure all auto checks have passed.

[gziz]

I'm wondering what's the difference between these two...

[jackgerrits]

Traditionally actual vs total is a matter of if the tokens were cached or not. But since we don't have caching atm they should be identical

[gziz]

Thanks @jackgerrits! I have updated the PR to use both actual_usage and total_usage.

[jackgerrits]

@gziz The original issue what that we're not using the returned value of the function where we add usage. I'd rather a smaller change where we just assign that value back to the appropriate variable. This change couples actual and total usage into the same function.

If we want to make all this simpler, we can implement__add__ and __iadd__:

• https://docs.python.org/3/reference/datamodel.html#object.__add__
• https://docs.python.org/3/reference/datamodel.html#object.__iadd__

That way we can update the code to simply be:

self._actual_usage += usage