fix(models): add GPT-6 Sol and Luna metadata

[Sun-GLiang]

Summary

• Display gpt-6-sol and gpt-6-luna as GPT-6 Sol and GPT-6 Luna while retaining their request IDs.
• Offer low, medium, high, xhigh, and max thinking levels on both OpenAI API and Codex OAuth connections. The pinned OpenAI Responses SDK drops GPT-6 none, so the UI no longer offers a nonfunctional Off choice.
• Route GPT-6 OpenAI API models through Responses so selected reasoning effort reaches the request.
• The Codex model inventory also advertises ultra for Sol. The current SDK drops that effort, so it is not offered until the request path can send and handle it.

Fixes #5632

Verification

• Core build and affected catalog/metadata/thinking suites: 61 passed.
• Runtime build; thinking and Responses wire suites: 66 passed. The wire test checks every offered level for both models on both connection paths against the actual request body.
• npm run lint, npm run format:check, npm run typecheck, and npm run check:model-metadata: passed.

UI screenshots

Captured with production UI components in temporary Storybook fixtures. The images are GitHub attachments and are not part of the commit. They show selection controls; no live model request was made.

Main task page

Before: raw model IDs	After: display names

Thinking selector after the fix, with Default and the five supported levels:

Model settings

Model catalog	Default thinking levels

AI use

• No generative tool made a substantive contribution
• Generative tooling made a substantive contribution

Tool(s) and scope: Codex (GPT-6) authored the metadata adaptation and regression tests. The commit includes a Generated-by trailer.

Checklist

• Tests cover the change and fail without it
• Lint, format, typecheck and the affected suites pass locally

Does this PR entail a change in behavior?

• Yes — described under Summary above
• No

[Astro-Han]

LGTM