feat: upgrade MiniMax default model to M3

[octo-patch]

Summary

Upgrade the MiniMax platform configuration in Langchain-Chatchat to use MiniMax-M3 as the default model. MiniMax-M1 / M1-40k / M2.5 are removed; M2.7 and M2.7-highspeed are kept as alternatives.

Changes

• Promote MiniMax-M3 to the default LLM (first entry in llm_models)
• Keep MiniMax-M2.7 and MiniMax-M2.7-highspeed as supported alternatives
• Remove older models: MiniMax-M2.5, MiniMax-M1, MiniMax-M1-40k
• embo-01 retained as the embedding model
• Route MiniMax embeddings through OpenAIEmbeddings (API-compatible)

How to use

MODEL_PLATFORMS:
  - platform_name: minimax
    platform_type: minimax
    api_base_url: https://api.minimax.io/v1
    api_key: YOUR_MINIMAX_API_KEY
    llm_models:
      - MiniMax-M3
      - MiniMax-M2.7
      - MiniMax-M2.7-highspeed
    embed_models:
      - embo-01

Why

MiniMax-M3 is the latest generation, with a 512K context window, up to 128K output, and image input support. Dropping the older M1/M2.5 entries keeps the default config focused on currently supported models.

Test plan

• Python files parse without errors
• Changes follow existing platform configuration patterns
• MiniMax API is OpenAI-compatible, so existing ChatOpenAI wrapper works without modification