feat: add MiniMax as LLM provider

[octo-patch]

Summary

Add MiniMax as a first-class LLM provider, following the same OpenAI-compatible pattern used by xAI, Moonshot, and DeepSeek.

• Provider registration in providers.ts with minimax: prefix routing and MINIMAX_API_KEY env var
• Model factory in model/llm.ts using ChatOpenAI with https://api.minimax.io/v1 base URL
• Model definitions in utils/model.ts: MiniMax M2.7 (1M context) and M2.7-highspeed
• Environment config updated in env.example and README.md

Models added

Model	Context	Use case
MiniMax-M2.7	1M tokens	Primary model for deep research
MiniMax-M2.7-highspeed	1M tokens	Fast variant for summarization

Files changed (9 files, 257 additions)

File	Change
src/providers.ts	Add MiniMax provider definition
src/model/llm.ts	Add MiniMax factory (ChatOpenAI + baseURL)
src/utils/model.ts	Add M2.7 and M2.7-highspeed models
env.example	Add MINIMAX_API_KEY
README.md	Add MiniMax API key mention
src/providers.test.ts	7 unit tests for provider registration + routing
src/utils/model.test.ts	5 unit tests for model definitions
src/model/llm.test.ts	5 unit tests for LLM factory
src/model/llm.integration.test.ts	3 integration tests for end-to-end pipeline

Test plan

• All 20 new tests pass (bun test)
• All 36 existing tests still pass (56 total)
• TypeScript type check passes (tsc --noEmit)
• Manual verification with MINIMAX_API_KEY set and /model command

How to test

# Set your MiniMax API key
export MINIMAX_API_KEY=your-key

# Run all tests
bun test

# Start Dexter and select MiniMax via /model command
bun start