feat: add MiniMax as LLM provider

[octo-patch]

Summary

Add MiniMax as a first-class LLM provider alongside OpenAI, Groq, OpenRouter, Gemini, Anthropic, and Ollama.

MiniMax offers the MiniMax-M2.7 and MiniMax-M2.7-highspeed models with 204K context windows via an OpenAI-compatible API, making it a drop-in alternative for crypto trading analysis.

Changes

• src/llm/__init__.py: Add minimax provider branch in get_llm() factory using ChatOpenAI with base_url=https://api.minimax.io/v1 and temperature=0.7 (MiniMax requires temperature in (0, 1])
• config.example.yaml: Add MiniMax configuration example and list minimax in provider comments
• README.md: Add MiniMax to supported providers list, add MINIMAX_API_KEY to env setup
• README_CN.md: Same updates for Chinese documentation
• tests/test_minimax_provider.py: 18 tests (12 unit + 3 config validation + 3 integration)

Usage

# Set API key
export MINIMAX_API_KEY=your-minimax-api-key

# config.yaml
model:
  name: "MiniMax-M2.7"
  provider: "minimax"

Test Plan

• 12 unit tests: provider instantiation, base URL, API key, model name, temperature, timeout, caching, backward compatibility with OpenAI/Groq/Anthropic
• 3 config tests: verify config.example.yaml, README.md, README_CN.md mention MiniMax
• 3 integration tests: MiniMax API connectivity, M2.7-highspeed model, temperature constraint
• Manual: configure config.yaml with provider: minimax and run backtest

All 18 tests pass.

---

MiniMax API docs: https://www.minimax.io/platform