support azure assistant api

[IANTHEREAL]

Why are these changes needed?

while using azure gpt assistant api, model in config_list would cause error: NotFoundError: Error code: 404 - {'error': {'code': '404', 'message': 'Resource not found'}}.

Following the azure tutorial - https://learn.microsoft.com/en-us/azure/ai-services/openai/how-to/assistant#create-an-assistant, we should remove model from config_list

Related issue number

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: 1 lines in your changes are missing coverage. Please review.

Comparison is base (cff9ca9) 39.44% compared to head (6fbe3b0) 65.07%.

Files	Patch %	Lines
autogen/agentchat/contrib/gpt_assistant_agent.py	83.33%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##             main    #1616       +/-   ##
===========================================
+ Coverage   39.44%   65.07%   +25.62%     
===========================================
  Files          56       56               
  Lines        5985     5990        +5     
  Branches     1333     1451      +118     
===========================================
+ Hits         2361     3898     +1537     
+ Misses       3431     1742     -1689     
- Partials      193      350      +157     

Flag	Coverage Δ
unittests	64.85% <83.33%> (+25.40%)	⬆️

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.

[IANTHEREAL]

@sonichi @gagb Hi, how should I add a azure assistant api testing? It requires an azure openai account.

[sonichi]

I made a few suggestions for future PR.
cc @gagb @olgavrou @ekzhu @BeibinLi @randombet @kinfey

For testing on GitHub, you can use config_list_from_json and add "api_type": ["2024-02-15-preview"] into the filter_dict.

[IANTHEREAL]

For testing on GitHub, you can use config_list_from_json and add "api_type": ["2024-02-15-preview"] into the filter_dict.

Ok. I'll see how to run the same test code twice using both OpenAI and Azure API separately.

[sonichi]

There is some problem. When users provide the model in the llm_config, it will be ignored.
Prior to this PR, when the model is provided in config_list, it will be ignored. Neither is comprehensive.
We can fix this issue later.

[gagb]

@gagb can you give it a try? I just fixed the code, I got the parameters of OpenAIWrapper wrong

Btw I did test it today. It worked with openai assistants, but I got stuck at trying to find an azure key that supported assistant API. The one I had access to did not :( I am working on getting it