docs: add MiniMax AI provider to Wave AI documentation

[faugustdev]

This PR adds documentation for the MiniMax AI provider to the Wave AI modes guide (docs/docs/waveai-modes.mdx).

Changes:

• Added MiniMax to the list of supported providers
• Added a new "MiniMax" section with configuration examples for three models:
  • MiniMax-M3 — frontier multimodal coding model with 1M context window
  • MiniMax-M2.7 — recursive self-improvement, professional office delivery
  • MiniMax-M2.7-highspeed — same performance as M2.7 with faster inference
• All configs use the OpenAI-compatible API endpoint (https://api.minimax.io/v1/chat/completions) and reference the MINIMAX_TOKEN secret

Example config:
{
"minimax-m3": {
"display:name": "MiniMax-M3",
"display:icon": "server",
"ai:apitype": "openai-chat",
"ai:model": "MiniMax-M3",
"ai:endpoint": "https://api.minimax.io/v1/chat/completions",
"ai:apitokensecretname": "MINIMAX_TOKEN",
"ai:capabilities": ["tools", "images", "pdfs"]
}
}

[CLAassistant]

All committers have signed the CLA.

[coderabbitai]

Caution

Review failed

An error occurred during the review process. Please try again later.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands.}