feat: add MiniMax as LLM provider

[octo-patch]

Summary

• Add MiniMax as a first-class LLM provider with MiniMax-M2.7, MiniMax-M2.5, and MiniMax-M2.5-highspeed models
• New MiniMaxEngine extending OpenAiEngine using OpenAI-compatible API at api.minimax.io/v1
• Full setup wizard integration: provider display, API key URL, recommended model, dynamic model fetching

Changes

File	Change
src/engine/minimax.ts	New MiniMax engine with think-tag stripping, low temperature for deterministic output
src/commands/config.ts	MINIMAX enum value, model list, API key URL, recommended model, validator
src/utils/engine.ts	Engine factory case for MiniMax
src/utils/errors.ts	Billing URL and recommended model for error recovery
src/utils/modelCache.ts	fetchMiniMaxModels() with /v1/models endpoint, cache integration
src/commands/setup.ts	Provider display name, listed in OTHER_PROVIDERS
README.md	Added minimax to provider list and config example
test/unit/minimax.test.ts	32 unit tests covering all integration points
test/unit/minimax-integration.test.ts	3 integration tests (M2.7, M2.5-highspeed, auth error)

Usage

oco config set OCO_AI_PROVIDER=minimax OCO_API_KEY=<your_minimax_api_key>

Or use the interactive setup wizard:

oco setup
# Select "Other providers..." → "MiniMax (M2.7, M2.5, fast inference)"

Test plan

• 32 unit tests pass (npm run test -- test/unit/minimax.test.ts)
• 3 integration tests pass with live MiniMax API (MINIMAX_API_KEY=... npm run test -- test/unit/minimax-integration.test.ts)
• Existing tests pass with no regressions (test/unit/config.test.ts, test/unit/removeContentTags.test.ts)
• Manual: oco setup → select MiniMax → generate commit message

[di-sukharev]

hello @octo-patch please merge master into your branch to solve the confict and i will merge it