Add MiniMax as first-class LLM provider (M2.7 + M2.7-highspeed, MINIMAX_API_KEY auto-detection, temp clamping)

[octo-patch]

Summary

This PR adds MiniMax as a first-class LLM provider for readmex.

Changes

src/readmex/config.py

• Added MINIMAX_API_KEY environment variable support: when set (and LLM_API_KEY is not set), it automatically configures the LLM provider with MiniMax defaults:
  • Base URL: https://api.minimax.io/v1
  • Model: MiniMax-M2.7
  • Individual settings (LLM_BASE_URL, LLM_MODEL_NAME) still take precedence

src/readmex/utils/model_client.py

• Added _is_minimax(base_url) — detects minimax.io / minimax.chat endpoints
• Added _get_effective_temperature() — clamps temperature to > 0.0 for MiniMax (the API rejects temperature=0)
• Added llm_is_minimax to get_current_settings() return dict

README.md / README_CN.md

• New "Using MiniMax as the LLM Provider" section under Configuration with env-var and config-file examples, plus a model table

tests/test_minimax_provider.py (19 tests, all passing)

• 6 config auto-detection unit tests
• 7 provider detection & temperature-clamping unit tests
• 6 integration tests with mocked HTTP

Quick start

export MINIMAX_API_KEY="<your key>"
export T2I_API_KEY="<your t2i key>"  # e.g. OpenAI for logo generation
readmex /path/to/your/project

Available MiniMax models

Model	Context	Notes
MiniMax-M2.7	204K	Recommended
MiniMax-M2.7-highspeed	204K	Faster, lower cost
MiniMax-M2.5	204K	Previous generation
MiniMax-M2.5-highspeed	204K	Previous generation, fast