[Feature]: Add Python Azure OpenAI support to traceai_openai

[VaishnavGunjari]

The current OpenAIInstrumentor only patches openai.OpenAI. The AzureOpenAI client is not instrumented, so Azure OpenAI calls are silently untraced.
This is a significant gap since Azure OpenAI is widely used in enterprise environments.

Proposed solution:
Extend OpenAIInstrumentor to wrap AzureOpenAI with Azure-specific span attributes (deployment_name, api_version, endpoint, gen_ai.system = "azure_openai").

I'd like to contribute this if it aligns with the roadmap. Happy to discuss before opening a PR.

[gh-raju]

I would like to take this by adding Azure OpenAI support to traceai_openai.

Proposed approach:

• Instrument AzureOpenAI or AsyncAzureOpenAI by reusing the existing OpenAI client wrapping, so all gen_ai.* span attributes stay identical across both providers.
• Handle the Azure specific pieces in attribute extraction: deployment name vs. model, api-version and endpoint.
• Mirror the existing OpenAI instrumentor tests and add a short quickstart note to the README.

Two scoping questions before I start:

1. Prefer a shared code path with a provider flag, or a separate AzureOpenAIInstrumentor?
2. How should deployment map onto gen_ai.request.model?

Could a maintainer assign this to me? I will branch from dev and keep the PR scoped to Azure support.