feat(llm): add MiniMax-M2.7 provider support

[octo-patch]

Summary

This PR adds MiniMax as a supported LLM provider, alongside the existing OpenRouter, OpenAI, Google Gemini, DeepSeek, and Grok providers.

MiniMax offers two chat models:

• MiniMax-M2.7 — Peak Performance, Ultimate Value (default)
• MiniMax-M2.7-highspeed — Same performance, faster and more agile

Because MiniMax exposes an OpenAI-compatible REST API, no new HTTP client code is needed — the existing _call_openai_compatible() method handles all requests.

Changes

File	What changed
app/services/llm.py	LLMProvider.MINIMAX enum value; provider config with base URL, default model, and fallback; API key mapping; model prefix detection/normalization for minimax/; MiniMax in auto-detection and fallback priority lists
app/config/api_keys.py	MINIMAX_API_KEY metaclass property (reads from MINIMAX_API_KEY env var or minimax.api_key config key)
app/utils/config_loader.py	MINIMAX_API_KEY, MINIMAX_BASE_URL, MINIMAX_MODEL environment variable mappings
app/routes/settings.py	MiniMax option in the LLM provider selector; API key, model, and base-URL settings items
env.example	MINIMAX_API_KEY, MINIMAX_MODEL, MINIMAX_BASE_URL documented

Configuration

LLM_PROVIDER=minimax
MINIMAX_API_KEY=your-key-here
MINIMAX_MODEL=MiniMax-M2.7
MINIMAX_BASE_URL=https://api.minimax.io/v1

How to test

1. Obtain a MiniMax API key from https://platform.minimax.io
2. Set LLM_PROVIDER=minimax and MINIMAX_API_KEY=<your-key> in backend_api_python/.env
3. Start the backend and run an AI analysis
4. Verify the provider appears in Settings → AI/LLM section of the UI