feat: add MiniMax as LLM provider

[octo-patch]

Summary

Add MiniMax as a supported LLM backend via its OpenAI-compatible API.

• Register MiniMax in backend2url and backend2model with https://api.minimax.io/v1
• Add auto model selection for MiniMax-M1 (supports up to 1M context window)
• Update config.ini with MiniMax documentation and model examples
• Add MiniMax to the LLM support table in both README.md and README_zh.md
• Add config example in setup instructions for both READMEs
• Add tests/test_minimax.py demo script, 20 unit tests, and 3 integration tests

Configuration Example

[llm.server]
remote_type = "minimax"
remote_api_key = "sk-xxxxxxxxxxxxx"
remote_llm_model = "MiniMax-M1"
remote_llm_max_text_length = 1000000

Test Plan

• 20 unit tests passing (pytest tests/test_minimax_unit.py -v)
• 3 integration tests passing with real MiniMax API (pytest tests/test_minimax_integration.py -v)
• Verify remote_type = "minimax" works end-to-end with python3 -m huixiangdou.main

Files Changed (7 files, ~433 additions)

File	Change
huixiangdou/services/llm.py	Add minimax to backend2url, backend2model, choose_model()
config.ini	Document minimax as supported provider
README.md	Add MiniMax to LLM support table + config example
README_zh.md	Add MiniMax to LLM support table + config example
tests/test_minimax.py	Demo script for MiniMax API
tests/test_minimax_unit.py	20 unit tests
tests/test_minimax_integration.py	3 integration tests