Improve efficiency for large organizations/collections in Azure DevOps

[sebdanielsson]

Tell us more.

I'm currently evaluating Renovate CLI for dependency updates on Azure DevOps Server. However, there are years of old repos that are no longer in use, and many of them are huge for many unnecessary reasons.

I had an idea to improve the speed by:

1. Turning on autodiscover
2. Turning off onboarding
3. Turning on optimizeForDisabled
4. Setting requireConfig to "required"

However, it seems like Renovate still clones every repo it has access to, even when it doesn't contain a renovate.json file and onboarding is off. Is there a reason for Renovate to clone the repo if onboarding is turned off? Or would you accept a PR to change this behavior? I'd be happy to try to tackle this, as it would probably improve the speed significantly for us.

Edit: Added point 4.

Edit 2: PR created - #39834