Convert ChatCompletionMessage to Dict after completion

[yiranwu0]

Why are these changes needed?

1. extract_text_or_function_call -> extract_text_or_completion_object. The extraction is now a dict or an completion object, changing the name for clarity.

2. Ensure the return from generate_oai_reply is a str or dict, which is a vulnerability.

Note: Remaining issues with groupchat will be fixed in a separate PR. (When selector is making a function call).

Related issue number

#721

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.

[codecov-commenter]

Codecov Report

Attention: 10 lines in your changes are missing coverage. Please review.

Comparison is base (a31b240) 26.72% compared to head (fe97bf2) 37.96%.

Files	Patch %	Lines
autogen/oai/client.py	37.50%	5 Missing ⚠️
autogen/agentchat/conversable_agent.py	0.00%	4 Missing ⚠️
autogen/agentchat/contrib/compressible_agent.py	0.00%	1 Missing ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             main     #791       +/-   ##
===========================================
+ Coverage   26.72%   37.96%   +11.24%     
===========================================
  Files          28       28               
  Lines        3768     3777        +9     
  Branches      854      900       +46     
===========================================
+ Hits         1007     1434      +427     
+ Misses       2687     2217      -470     
- Partials       74      126       +52     

Flag	Coverage Δ
unittests	37.91% <23.07%> (+11.24%)	⬆️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[afourney]

I think this is also related to #765