Add MCP cookbook for OpenAI-Gemini compatibility

[Ramachetan]

This PR adds a new cookbook example demonstrating how to create a Chainlit application that works seamlessly with both Google Gemini and OpenAI models using a single codebase.

Changes

Created new folder mcp-openai-gemini with a dual-compatible implementation using the OpenAI SDK
Added comprehensive README.md explaining how to configure for either OpenAI or Gemini models
Included environment setup instructions and configuration examples

Why this matters

This cookbook provides developers with a practical example of how to build MCP Client applications that can switch between LLM providers with minimal code changes, leveraging Google AI Studio's OpenAI SDK compatibility layer.

[dkopljar27]

It would be great to also add support for AzureOpenAI...
Add import> from openai import AsyncAzureOpenAI

Init AzureClient>
client = AsyncAzureOpenAI( api_key=API_KEY, api_version=API_VERISON, azure_endpoint=ENDPOINT )

Add step for empty first chunk (due to e.g. content filtering)

 async for chunk in stream_resp:
    if not chunk.choices:
        continue
    delta = chunk.choices[0].delta
    if delta and delta.content:
        await msg.stream_token(delta.content)

[Ramachetan]

I can work on it but I don't have access to azure open ai right so, can't really test it