Integrate Inference SDK

[kaylieee]

Purpose

• Support is added for creation and use of Azure Inference clients for assistants of type ChatAssistantClient (chat completion API based) in the Assistant Tool and SDK middleware library
• Add new AIClient and AsyncAIClient classes to improve extensibility making adding support for new models easier
• Add new AIClientConfig class to manage and store configurations of created AI clients
• Version updated to 0.4.4-alpha

Does this introduce a breaking change?

AIClientFactory now returns ai clients of type OpenAIClient, AzureOpenAIClient, or AzureInferenceClient or their asynchronous counterparts. To interface with the underlying client, you can use property ai_client.

[X] Yes
[ ] No

Pull Request Type

What kind of change does this Pull Request introduce?

[ ] Bugfix
[X] Feature
[ ] Code style update (formatting, local variables)
[ ] Refactoring (no functional changes, no api changes)
[ ] Documentation content changes
[ ] Other... Please describe:

How to Test

• Get the code

git clone [repo-address]
cd [repo-name]
git checkout [branch-name]
npm install

• Test the code

Run pytests in sdk/azure-ai-assistant/test

Other Information

• Please install new version azure_ai_assistant-0.4.4a1-py3-none-any.whl for middleware SDK from this release. See the README for installation details.
• Azure Inference clients use OpenAI threads for the purpose of chat memory in the tool. To use Azure Inference clients with the Assistant Tool, you must have an OpenAI resource.