Add AudioBytes Class and OpenAI Model Support for Audio Input

[cto-new]

Summary

This PR introduces the AudioBytes class to represent audio input bytes and enhances OpenAI chat model support to accept audio (WAV) input as part of user messages. It also ensures consistent type annotations and casting throughout the codebase to maintain compatibility with static type checkers and newer Python standards.

Details

• Adds AudioBytes class in chat_model/message.py for audio (WAV) input and validates MIME type.
• Incorporates AudioBytes into UserMessageContentBlock for message flexibility.
• Serializes AudioBytes correctly in OpenAI chat model and allows their use in user content.
• Updates OpenAI user message conversion and formatting for audio blocks.
• Exports AudioBytes from the main module for external availability.
• Updates all typing.cast() usage to use string literal type expressions as required by PEP 695 and enforced by tooling, ensuring correct syntax and compatibility throughout the repo.
• Fixes Python formatting, and ensures function signatures and test type hints match new standards.

[cto-new]

Engine is processing your review feedback and will update the code shortly...

[cto-new]

Engine is processing your review feedback and will update the code shortly...