[aksg87]

Description

Summary

• OpenAI shuts down GPT-4o, GPT-4o mini, GPT-4.1 and GPT-4.1 mini on
  January 14, 2027 and recommends gpt-6-luna as the replacement
  (deprecations). No
  built-in pattern matches gpt-6* IDs, so on main
lx.extract(model_id="gpt-6-luna") raises
  InferenceConfigError: No provider registered for model_id='gpt-6-luna'.
• Adds ^gpt-6 next to ^gpt-5 in the OpenAI provider patterns. Built-in
  routing for existing model IDs is unchanged, and gpt-oss* still routes to
  Ollama. Updates the routing lists in the bundled usage skill docs.

Bug fix

How Has This Been Tested?

Tests

• New: RegistryTest.test_openai_patterns_cover_gpt6_models (gpt-6-luna,
  gpt-6.1-sol) and a gpt6_luna case in
  OpenAIRoutingExtractionTest.test_extract_with_environment_key
  (end-to-end lx.extract with a mocked client and an environment API key).
  Both fail on main.
• pytest -m 'not live_api and not requires_pip and not integration':
  985 passed, 25 deselected.
• isort 5.13.2 and pyink 24.3.0 checks are clean; pylint is clean on the
  changed files.
• Live: this PR's live-api-tests job runs the existing live tests, none of
  which call a GPT-6 model. Move OpenAI live tests off retiring GPT-4o models #601 exercises gpt-6-luna routing live.
  Pending; added once this run finishes.

Checklist

• I have made any needed documentation changes.
• I have added tests, or I have ensured existing tests cover the changes.
• I have followed Google's Python Style Guide and ran pylint over the
  affected code.