feat: support reasoning field

[kwaa]

closed #258

[gemini-code-assist]

Summary of Changes

Hello @kwaa, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request enhances the system's ability to handle and utilize AI model reasoning by introducing explicit support for a new reasoning field. This change involves updating core data interfaces and modifying the text streaming and generation logic to correctly process and prioritize this new field. The primary goal is to leverage more detailed chain-of-thought information from AI models, improving the richness of generated responses and aligning with evolving model outputs.

Highlights

• New 'reasoning' field support: Introduced a new reasoning field in relevant interfaces (StreamTextChunkResult.choices.delta and AssistantMessage) to capture AI model's chain of thought, with a reference to an OpenAI cookbook article.
• Prioritization in stream processing: Updated the streamText utility to detect and prioritize the new reasoning field over the existing reasoning_content when processing streamed text chunks, ensuring the most relevant reasoning is captured.
• Dynamic message construction: Modified the final message construction in streamText to dynamically include either the reasoning or reasoning_content field based on which one was present and prioritized during streaming.
• Updated text generation logic: The generateText function now prioritizes the reasoning field when extracting reasoningText from assistant messages.
• Code cleanup: Removed several commented-out and unused lines of code related to telemetry and usage tracking in the streamText implementations.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request adds support for a reasoning field, as an alternative to reasoning_content, to handle different AI provider responses. The changes are applied consistently across various packages, updating interfaces and processing logic. The implementation is mostly solid, but I've identified an opportunity to refactor and simplify the logic for handling the new reasoning field in the streaming functions, which would improve maintainability and robustness. Additionally, several pieces of commented-out code have been cleaned up, which is a welcome improvement.