Added multi-modal support

[robgruen]

This pull request adds support for multimodal prompts (combining images and text) to the TypeChat project, introduces an example project demonstrating this capability, and updates documentation and solution files accordingly. The main focus is enabling vision-capable models (like GPT-4o) to process images alongside text and translate what they see into strongly typed objects.

Key changes include:

Multimodal Support and Example:

• Added a new example project, Multimodal, which demonstrates how to send images to a vision-capable model and receive a strongly typed description (ImageResponse). This includes schema definition and project setup. [1] [2]
• Integrated the Multimodal example into the solution (TypeChat.sln) and updated build configurations to support it. [1] [2]

Documentation Updates:

• Updated README.md to document the new multimodal prompt feature, describe how to use images in prompts, and mention the new OpenAI Responses API support.
• Added a reference to the new Multimodal example in the list of project examples in README.md.