feat: add MiniMax as LLM provider with M2.5 models

[octo-patch]

Summary

• Add MiniMax as a 4th LLM provider alongside OpenRouter, Groq, and HacxGPT
• MiniMax offers an OpenAI-compatible API at https://api.minimax.io/v1 with 204K context window
• Two models: MiniMax-M2.5 (general) and MiniMax-M2.5-highspeed (optimized for speed)

Changes

File	Change
config_manager.py	Add minimax provider entry with name, URL, and 2 models
chat.py	Add MiniMax API endpoint routing in chat_request()
models.py	Add MiniMax-M2.5 and MiniMax-M2.5-highspeed ModelSpec entries (204K context)
main.py	Update About, Links, and Disclaimer sections to include MiniMax
README.md	Add MiniMax to provider tables, prerequisites, FAQ, and project structure
tests/test_minimax_provider.py	31 tests (25 unit + 3 integration + 3 edge case)

How to use

1. Get an API key at platform.minimaxi.com
2. Select minimax as provider in Settings (option 2)
3. Choose a model: MiniMax-M2.5 or MiniMax-M2.5-highspeed
4. Enter your API key and start chatting

Test plan

• All 31 tests pass (python -m unittest tests/test_minimax_provider.py -v)
• Unit tests cover config, model registry, chat request routing, error handling
• Integration tests verify real API calls with MiniMax-M2.5 and M2.5-highspeed
• Manual: select minimax provider in Settings menu and chat