feat(core): add MiniMax provider support

[octo-patch]

Why we need this PR?

Acontext needs configuration support for MiniMax models through its existing compatible LLM adapters.

Describe your solution

• Register minimax as an OpenAI-compatible LLM SDK option.
• Include LLM_SIMPLE_MODEL in CLI-generated environment files while preserving the existing defaults.
• Document MiniMax-M3 and MiniMax-M2.7 capabilities, pricing tiers, and global and China OpenAI/Anthropic endpoints.

Implementation Tasks

• Add provider registry and schema support.
• Generate and test the configured simple model in CLI environment files.
• Document the complete model and endpoint matrix.

Impact Areas

Which part of Acontext would this feature affect?

• Client SDK (Python)
• Client SDK (TypeScript)
• Core Service
• API Server
• Dashboard
• CLI Tool
• Documentation
• Other

Checklist

• Open this pull request against the dev branch.
• All relevant tests pass.
• Tests are added or modified to cover code changes.

Test plan

• go test ./... from src/client/acontext-cli
• LLM_API_KEY=test-key uv run pytest -q tests/llm/test_provider_registry.py from src/server/core
• uv run python -m py_compile acontext_core/schema/config.py acontext_core/llm/complete/__init__.py tests/llm/test_provider_registry.py from src/server/core
• ./node_modules/.bin/next build from docs
• Captured request paths for all four documented region/protocol endpoint combinations.