Avoid using get_info on bad elements

[GiudGiud]

they could trigger errors themselves

this was preventing a friend from looking at how many bad Jacobian elements they had.
And there were very many (1500), from a prism21 mesh generated in an external professional mesh generator. So this either means we have an error computing the Jacobian of prism21 elements in libmesh (I will tag the relevant people once holidays are past) or the mesh generation software can be made to create bad prism21s

refs #25278

[moosebuild]

Job Documentation on 688c1c9 wanted to post the following:

View the site here

This comment will be updated on new commits.

[moosebuild]

Job Coverage on 688c1c9 wanted to post the following:

Framework coverage

	4d71af	#26410 688c1c
	Total	Total	+/-	New
Rate	85.07%	85.07%	+0.00%	100.00%
Hits	98385	98387	+2	7
Misses	17265	17265	-	0

Diff coverage report

Full coverage report

Modules coverage

Coverage did not change

Full coverage reports

Reports

• framework
• chemical_reactions
• combined
• contact
• electromagnetics
• external_petsc_solver
• fluid_properties
• fsi
• functional_expansion_tools
• geochemistry
• heat_transfer
• level_set
• misc
• navier_stokes
• optimization
• peridynamics
• phase_field
• porous_flow
• ray_tracing
• rdg
• reactor
• richards
• scalar_transport
• solid_properties
• stochastic_tools
• tensor_mechanics
• thermal_hydraulics
• xfem

This comment will be updated on new commits.

[GiudGiud]

Thanks for the review!

[roystgnr]

What's the stack trace on the errors from get_info()? Are we simply hitting failures when computing JxW in Elem::volume()? We probably ought to be doing that in a try/catch block, so we can return NaN rather than throw an error when the volume isn't well-defined...

IIRC I checked the ExodusII docs carefully when adding Prism21 import, but it's possible I made a mistake. Can you get me a minimal failing mesh?

[GiudGiud]

yes.
That works for me too. get_info was nice in that log, but what we have now is enough. I dont mind making another PR if libmesh changes get_info() to survive awful elements.

Sending to slack in 5

[roystgnr]

libMesh/libmesh#3758 should survive awful elements; at least it survives the particular awful element in the unit test.