feat: add MiniMax provider support

[ximiximi423]

Summary

• Add MiniMax as a new model provider with OpenAI-compatible API integration
• Support two models: MiniMax-M2.5 (default) and MiniMax-M2.5-highspeed
• Both models support 204,800 tokens context window

Supported Models

Model ID	Description
MiniMax-M2.5	Peak Performance. Ultimate Value. Master the Complex
MiniMax-M2.5-highspeed	Same performance, faster and more agile

Changes

• Add minimax flavor to LlmProvider enum in shared models
• Add MiniMax provider creation using @ai-sdk/openai-compatible SDK (OpenAI-compatible API)
• Add MiniMax to CLI provider configuration with default API key env var, base URL, and model
• Add MiniMax to desktop app (Rowboat X) settings dialog with provider config
• Add MiniMax to onboarding model discovery list
• Add MINIMAX_API_KEY environment variable support in .env.example and docker-compose.yml

API Details

• Endpoint: https://api.minimax.io/v1 (OpenAI-compatible)
• Auth: Bearer token via MINIMAX_API_KEY

API Documentation

• OpenAI-compatible API
• Anthropic-compatible API

Test Plan

• Verify MiniMax provider appears in CLI model configuration
• Verify MiniMax provider appears in desktop app settings
• Test model connection with MiniMax API key
• Verify chat completion with MiniMax-M2.5 model

[vercel]

@ximiximi423 is attempting to deploy a commit to the RowBoat Labs Team on Vercel.

A member of the Team first needs to authorize it.