In many of the sync jobs, string replacements transform upstream into downstream or midstream into downstream.

But instead of ugly sed code, wrapped by bash commands, inside groovy, we could simply use groovy directly.

Here's an example of nested regex:

https://github.com/redhat-developer/codeready-workspaces-operator/blob/master/operator-metadata.Jenkinsfile#L127-L130

and a cleaner example of just using groovy directly:

https://github.com/redhat-developer/codeready-workspaces-operator/blob/master/operator.Jenkinsfile#L135-L153

Where possible, we should move to this cleaner and less slash-escaped-to-death code.