feat: add MiniMax M2.7/M2.5 provider support via minimax/ prefix

[octo-patch]

Summary

This PR adds first-class MiniMax LLM provider support to llmgraph via a minimax/ model prefix, routing through LiteLLM's OpenAI-compatible API path.

What changed

• llmgraph/library/llm.py — added _is_minimax_model() and _resolve_model_config() helpers; updated make_call() to auto-route minimax/<model> to openai/<model> with MiniMax's API base URL and clamped temperature
• llmgraph/library/consts.py — added minimax_api_base, minimax_default_model, and minimax_model_prefix constants
• tests/test_llm_minimax.py — 22 tests (9 unit _is_minimax_model, 11 unit _resolve_model_config, 3 integration tests for make_call)
• README.md — added MiniMax usage example

Usage

export MINIMAX_API_KEY=your_api_key_here
llmgraph machine-learning "https://en.wikipedia.org/wiki/Artificial_intelligence" \
  --levels 3 --llm-model minimax/MiniMax-M2.7

Supported models: MiniMax-M2.7, MiniMax-M2.7-highspeed, MiniMax-M2.5, MiniMax-M2.5-highspeed (all 204K context).

Provider-specific handling

Behaviour	Detail
Base URL	Auto-set to https://api.minimax.io/v1 when using minimax/ prefix
Auth	Set MINIMAX_API_KEY env var; LiteLLM picks it up automatically
Temperature	Clamped to (0.0, 1.0] — MiniMax rejects temperature=0
Model routing	minimax/MiniMax-M2.7 → openai/MiniMax-M2.7 via LiteLLM OpenAI-compat path

Test results

22 passed in 8.29s