feat: add MiniMax provider support

[octo-patch]

Summary

• Add MiniMax chat model support via OpenAI-compatible API
• Add model shortcuts: minimax (→ MiniMax-M2.7) and minimax-highspeed (→ MiniMax-M2.7-highspeed)
• Add unit tests for MiniMax provider detection and model aliases

Usage

# Set your MiniMax API key
export MINIMAX_API_KEY=your-key-here

# Use MiniMax-M2.7 (default)
promptr -m minimax -p "Refactor this function" src/myfile.js

# Use MiniMax-M2.7-highspeed
promptr -m minimax-highspeed -p "Add error handling" src/myfile.js

# Use full model name directly
promptr -m MiniMax-M2.7 -p "Add unit tests" src/myfile.js

Implementation Details

• Detects MiniMax models by checking if the model name starts with minimax (case-insensitive)
• Uses MINIMAX_API_KEY environment variable and https://api.minimax.io/v1 base URL
• Uses the modern tools API format (MiniMax does not support the legacy functions API)
• Places system messages before user messages (required by MiniMax's API)
• Clamps temperature to the valid range (0.0, 1.0] (MiniMax does not accept temperature=0)
• Custom base URL override via MINIMAX_BASE_URL environment variable

API Reference

• Chat (OpenAI Compatible): https://platform.minimax.io/docs/api-reference/text-openai-api