feat: add MiniMax provider support with M2.7 default model

[octo-patch]

Summary

Add MiniMax as a new LLM provider via OpenAI-compatible API, with the latest M2.7 as the default model.

Changes

• Add MiniMax provider with OpenAI-compatible API integration
• Include MiniMax-M2.7 and MiniMax-M2.7-highspeed as primary models
• Retain MiniMax-M2.5 and MiniMax-M2.5-highspeed as alternatives
• Configure model capabilities (204K context, pricing, tool support)
• Add provider settings (API key, help links)

Models

Model	Context	Input	Output
MiniMax-M2.7 (default)	204K	$0.30/M	$1.20/M
MiniMax-M2.7-highspeed	204K	$0.60/M	$2.40/M
MiniMax-M2.5	204K	$0.30/M	$1.20/M
MiniMax-M2.5-highspeed	204K	$0.60/M	$2.40/M

Why

MiniMax-M2.7 is the latest flagship model with enhanced reasoning and coding capabilities.

Testing

• Model capabilities correctly defined with proper typing
• Fallback handler routes unknown MiniMax models to M2.7 default
• OpenAI SDK integration with MiniMax base URL verified