Fix run-pass telemetry decorator placement

[Sylvester Kaczmarek (sylvesterkaczmarek)]

Describe your changes

Addresses the run-pass decorator issue identified in #2516.

RunPassCommand currently applies @action to the class itself, while the other CLI commands apply it to their run() methods. The action decorator wraps callables and therefore replaces the class binding with a wrapper function, recording construction rather than the actual command execution.

This change keeps RunPassCommand as a class and applies @action to run() consistently with the rest of the CLI.

Tests

Adds regression coverage verifying that RunPassCommand remains a class and that its run() method is wrapped by the telemetry decorator.

[azure-pipelines]

Azure Pipelines:
There may be pipelines that require an authorized user to comment /azp run to run.

[Copilot]

🟢 Approval recommended

The change aligns run-pass with existing CLI telemetry patterns and includes a focused regression test validating the intended decorator placement.

Pull request overview

This PR fixes Olive CLI telemetry for the run-pass command by moving the @action decorator from the RunPassCommand class (which unintentionally wraps the class into a function) onto the run() method, aligning it with the established pattern used by other CLI commands.

Changes:

• Removed @action from RunPassCommand’s class definition so the argparse binding remains a class (not a wrapper function).
• Added @action to RunPassCommand.run() so telemetry records actual command execution.
• Added a regression test to ensure RunPassCommand remains a class and that run() is wrapped (via functools.wraps / __wrapped__).

File summaries

File	Description
olive/cli/run_pass.py	Moves @action from the command class to the run() method to correctly capture execution telemetry and preserve class binding.
test/cli/test_run_pass_action.py	Adds regression coverage asserting the command is still a class and the run() method is telemetry-wrapped.

Review details

• Files reviewed: 2/2 changed files
• Comments generated: 0
• Review effort level: Lite

---

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