fix: backfill azure/gpt-5.3-codex in provider model lists

[dzianisv]

Issue for this PR

Closes #106

Type of change

• Bug fix
• New feature
• Refactor / code improvement
• Documentation

What does this PR do?

The Azure provider list can miss azure/gpt-5.3-codex when the local models.dev payload is older than the latest Azure catalog. This change backfills that model in the core provider normalization path so the provider list, model picker, and opencode models azure still expose it even when metadata lags.

It also adds regression coverage for that stale-metadata case and documents the local MCP idle cleanup behavior that this fork carries.

How did you verify your code works?

• Ran bun test test/provider/provider.test.ts test/provider/transform.test.ts
• Ran bun run typecheck
• Verified the CLI lists the model against the stale fixture with:
  OPENCODE_MODELS_PATH=/Users/engineer/workspace/opencode/packages/opencode/test/tool/fixtures/models-api.json AZURE_RESOURCE_NAME=test-resource AZURE_API_KEY=test-api-key bun run --cwd packages/opencode --conditions=browser src/index.ts models azure | rg '^azure/gpt-5\\.3-codex$'

Screenshots / recordings

N/A. This is not a UI change.

Checklist

• I have tested my changes locally
• I have not included unrelated changes in this PR

[github-actions]

This PR doesn't fully meet our contributing guidelines and PR template.

What needs to be fixed:

• PR description is missing required template sections. Please use the PR template.

Please edit this PR description to address the above within 2 hours, or it will be automatically closed.

If you believe this was flagged incorrectly, please let a maintainer know.

[github-actions]

This pull request has been automatically closed because it was not updated to meet our contributing guidelines within the 2-hour window.

Feel free to open a new pull request that follows our guidelines.