feat(models): add MiniMax-M3 and MiniMax-M2.7 to recommended models

[octo-patch]

Reason: Add MiniMax-M3 and MiniMax-M2.7 to Strix's recommended model metadata and provider documentation with correct token contexts and regional endpoints.

Changes

• Add openai/MiniMax-M3 (1,000,000-token context) and openai/MiniMax-M2.7 (204,800-token context) to RECOMMENDED_MODEL_NAMES in strix/config/models.py so they are recognized as recommended frontier models.
• Add a minimax entry to FRONTIER_MODEL_FAMILIES so bare and prefixed MiniMax model names are matched by is_recommended_or_frontier_model.
• Add a new docs/llm-providers/minimax.mdx provider guide documenting both models with their correct context windows and both regional endpoints (global api.minimax.io and China api.minimaxi.com).
• Register the MiniMax guide in docs/docs.json navigation and link it from the providers overview and the README recommended-models list.

Checks

• uv run ruff check strix/config/models.py tests/test_models.py — all checks passed
• uv run ruff format --check strix/config/models.py tests/test_models.py — already formatted
• uv run pytest tests/test_models.py -q — 70 passed

[greptile-apps]

Correct the provider routing description

The openai/ prefix routes this configuration through Strix's native OpenAI provider, not its LiteLLM fallback. Calling it a LiteLLM prefix misdirects users troubleshooting the documented MiniMax setup.

Suggested change

	- MiniMax API is fully OpenAI-compatible, so it works via the `openai/` LiteLLM prefix with `LLM_API_BASE`
	- MiniMax API is fully OpenAI-compatible, so it works via the native `openai/` provider prefix with `LLM_API_BASE`

Knowledge Base Used: Configuration and Telemetry

Prompt To Fix With AI

This is a comment left during a code review.
Path: docs/llm-providers/minimax.mdx
Line: 53

Comment:
**Correct the provider routing description**

The `openai/` prefix routes this configuration through Strix's native OpenAI provider, not its LiteLLM fallback. Calling it a LiteLLM prefix misdirects users troubleshooting the documented MiniMax setup.

```suggestion
- MiniMax API is fully OpenAI-compatible, so it works via the native `openai/` provider prefix with `LLM_API_BASE`
```

**Knowledge Base Used:** [Configuration and Telemetry](https://app.greptile.com/strix-org-3/-/custom-context/knowledge-base/usestrix/strix/-/docs/telemetry-and-config.md)

---

For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.

Note: If this suggestion doesn't match your team's coding style, reply to this and let me know. I'll remember it for next time!

[greptile-apps]

Greptile Summary

This PR adds MiniMax-M3 and MiniMax-M2.7 to recommended/frontier model metadata and documents MiniMax configuration.

• Registers both model identifiers and the MiniMax frontier family.
• Adds tests for exact and bare MiniMax model-name matching.
• Adds a MiniMax provider guide, navigation entry, overview card, and README recommendation.

Confidence Score: 4/5

The PR appears safe to merge, with one non-blocking documentation correction needed for the description of MiniMax provider routing.

The added model metadata and tested classification paths are internally consistent, while the provider guide incorrectly labels the native openai/ provider prefix as a LiteLLM prefix.

Files Needing Attention: docs/llm-providers/minimax.mdx

Important Files Changed

Filename	Overview
strix/config/models.py	Adds exact recommended MiniMax identifiers and family-prefix classification without an identified runtime defect.
tests/test_models.py	Extends model-classification coverage for both documented identifiers and bare MiniMax family names.
docs/llm-providers/minimax.mdx	Adds setup, endpoint, and model guidance, but inaccurately describes the openai/ routing prefix as LiteLLM.
docs/docs.json	Registers the new provider guide in documentation navigation.
docs/llm-providers/overview.mdx	Adds MiniMax to the provider table and guide cards.
README.md	Adds MiniMax-M3 to the recommended-model list and provider overview text.

Prompt To Fix All With AI

### Issue 1
docs/llm-providers/minimax.mdx:53
**Correct the provider routing description**

The `openai/` prefix routes this configuration through Strix's native OpenAI provider, not its LiteLLM fallback. Calling it a LiteLLM prefix misdirects users troubleshooting the documented MiniMax setup.

```suggestion
- MiniMax API is fully OpenAI-compatible, so it works via the native `openai/` provider prefix with `LLM_API_BASE`
```

---

For each issue above, determine whether it is valid and should be fixed. If so, fix it directly.

_{Reviews (1): Last reviewed commit: "feat(models): add MiniMax-M3 and MiniMax..." | Re-trigger Greptile}