Consolidate XL clusters into normal regional clusters

[kylos101]

Create a new werft job or alter the existing one to support a new optional parameter workspace-cluster-new. XL clusters (require alternate manifests and exclusive admission constraints and scoring) as per this PR. We've been deploying them from a branch, and then setting admission constraints and score manually.

- [ ] @atduarte we need to turn this into an epic, a super tiny one, it'll have an ops change and a code change. Let's talk on Monday.

Traffic shifts have also been manual, and should be automated too.

• Need support for workspace classes, so we can add node labels, so that pods can be scheduled to be isolated from unleashed versus other
• ws-daemon config will vary by nodepool (might be naive)
• ws-manager pod yaml will vary by nodepool (might be naive)
• [ ] workspace-templates will vary by nodepool (might be naive)

This was done with the introduction of workspace classes in #9705

[atduarte]

@kylos101 this is progress right?

[kylos101]

Closing as this is resolved.