Velero policy for AI generated PRs

[Lyndon-Li]

We see increasing involvement of AI activities to Velero project, e.g., some PRs are generated purely by AI.
We need to create a policy for guidance of contributors/maintainers of how AI should be used to Velero project.

[Lyndon-Li]

cc @vmware-tanzu/velero-maintainers

@weshayutin Not sure if the title of this issue could cover all the requirements, if not, feel free to modify it.

[sseago]

The scope of this should probably include both AI-generated PRs and PRs generated by developers who used AI in the generation of the PR content. For example, I use Claude to generate some scaffolding, or to add a new server arg "foo", then I modify the code to make use of that new server arg, then I get claude to generate some unit tests, but then I modify those to actually test the things I need tested, etc. That's a very different category from a PR that was entirely AI-generated. This policy should probably address both.

[weshayutin]

/assign weshayutin