Use Codex-compatible model for daily CLI performance workflow

[Copilot]

The Daily CLI Performance Agent failed because openai/gpt-5.4 rejected Codex’s custom tool type.

• Workflow configuration: Switch the agent to openai/gpt-5.3-codex.
• Generated workflow: Regenerate the corresponding lock file.
• Regression coverage: Add a focused test ensuring the workflow retains a Codex-compatible model.

engine:
  id: codex
model: openai/gpt-5.3-codex

• Fixes [aw] Daily CLI Performance Agent failed #58825

[pelikhan]

@copilot use copilot/...

[Copilot]

🟢 Approval recommended

The source, generated workflow, and focused regression test are consistent with the reported failure and introduce no security regression.

Pull request overview

Updates the daily CLI performance workflow to use a Codex-compatible model, resolving #58825.

Changes:

• Switches to openai/gpt-5.3-codex.
• Regenerates the compiled workflow.
• Adds regression coverage for engine/model configuration.

File summaries

File	Description
.github/workflows/daily-cli-performance.md	Updates the configured model.
.github/workflows/daily-cli-performance.lock.yml	Propagates the model through generated jobs.
pkg/workflow/prompts_test.go	Verifies the Codex engine and compatible model.

Review details

• Files reviewed: 3/3 changed files
• Comments generated: 0
• Review effort level: Balanced

---

💡 Add a code-review agent skill for context-aware, tailored reviews. Learn more in the docs.

[Copilot]

@copilot use copilot/...

Updated the workflow to use model: copilot/gpt-5.3-codex with model-provider: github, regenerated the lock file, and updated the regression test. Commit: b7b6e2d.

[github-actions]

🎉 This pull request is included in a new release.

Release: v0.88.5