[JENKINS-55016] Disable flaky UtilTest.testDeleteRecursive_onWindows

[batmat]

It is getting in the way of other in flight PRs, so we should disable to possibly take the time more serenely to fix this flakiness.

See JENKINS-55016.

For instance:

• Add periods to sentences #3781 => text only modification, makes this fail
• [WIP] Bump to parent pom 1.51 #3784
• Bump to latest 1.50 parent pom release #3783 where @jglick even already advised to disable this one
• ...

Proposed changelog entries

• N/A

Submitter checklist

• JIRA issue is well described
• Changelog entry appropriate for the audience affected by the change (users or developer, depending on the change). Examples
  * Use the Internal: prefix if the change has no user-visible impact (API, test frameworks, etc.)
• Appropriate autotests or explanation to why this change has no tests
• For dependency updates: links to external changelogs and, if possible, full diffs

Desired reviewers

[batmat]

I will likely merge this one either later today (at least 4 hours), or tomorrow morning if I do not get pushback in the meantime.

I'm shortening the timeline a bit compared to usual given this has been blocking a lot of PRs, and given the ongoing issue with core rebuilds and agents' consumption, theses PRs make things even worse (because author will understandably retrigger their PR build to get a green build...)

[jvz]

I have a todo item to refactor this test to be more reliable. I'll try to contribute a fixed version back within a week or two.

[batmat]

Falling on poorManPerformanceTestBed only now. Going to merge and will probably file a similar PR for this other one tomorrow likely.

[batmat]

Filed #3788 for analyzing the poorManPerformanceTestBed issue already.