Set verify_holes to false in RGMB FlexiblePatternGenerator mesh subgenerator calls

[shikhar413]

Reason

Design

Impact

refs #28316

[shikhar413]

@milljm this PR should address the test failure mentioned in #28231

[milljm]

Keeping an eye on it: https://civet.inl.gov/job/2370876/

[roystgnr]

I assume the problem here is downstream of our failure to fix libMesh/libmesh#3497 ?

But I'd love to be reassured that that's more than an assumption, that the point (x,y,z)=( 2.04665, 1.11022e-16, 0) on hole boundary but outside outer boundary has been checked and definitely isn't outside the outer boundary.

It's probably a good idea to turn off this check either way, since it is going to give us false positives until we can fix it, but if we let a bad test case slip through then we'll find ourselves stuck getting true positives after we fix it.

[moosebuild]

Job Documentation on e80e72d wanted to post the following:

View the site here

This comment will be updated on new commits.

[shikhar413]

I assume the problem here is downstream of our failure to fix libMesh/libmesh#3497 ?

@roystgnr yes these changes are needed to get the Conda Intel Mac tests pass, which was triggered when adding the feature in this PR - #28231. Basically the idea here is to delete the outermost layer created by the RGMB mesh generators and triangulate this layer instead, which will facilitate stitching of assembly mesh structures without having any hanging nodes. Since we're just triangulating a layer that was previously deleted through the mesh subgenerators, we can be sure that the hole (layers not deleted in assembly) is contained within the outer boundary (outermost assembly boundary). These checks happen through the mesh subgenerators that are called before this deletion and triangulation step.

[shikhar413]

@milljm looks like the devel tests that were previously failing now pass

[GiudGiud]

the holes are placed programmatically, and the flexible pattern sub-generator already checks that they dont overlap iirc

this should be safe to do for now though we can re-enable it if we find that users have a way to make the hole placement invalid