[Feature Request] Support Local AI Models (Ollama / LM Studio) for Commit Message Generation #2430
Description
Summary
Add support for using local AI models (via Ollama or LM Studio) to generate commit messages directly in Fork.
Motivation
• Current AI-assisted commit generation typically relies on cloud services (e.g., ChatGPT, Claude).
• Many developers prefer not to send diffs or code externally due to privacy, security, or compliance concerns.
• Local model inference (Ollama, LM Studio) allows AI-powered commit messages while keeping all data on-device.
Proposal
• Allow Fork to connect to a local model provider (Ollama/LM Studio).
• Fork could send the staged diff to the local API and receive a suggested commit message.
• Users can then edit/approve before committing.
Benefits
• Privacy & security (code never leaves the machine).
• Performance (local inference is fast with optimized models).
• Flexibility (developers can pick or fine-tune models).