Add Polygon/Polyhedron NSIDED/NFACED ExodusII output

[ConnorOuellette]

Summary

This adds ExodusII output support for libMesh’s arbitrary polygon and polyhedron elements:

• writes C0POLYGON elements as Exodus NSIDED blocks
• writes C0POLYHEDRON elements as Exodus NFACED blocks
• emits the associated face block, face-node connectivity, face counts, element-face connectivity, and element face counts needed for NFACED output

Notes

• If using ParaView, the default IOSS Reader does not support polygonal/polyhedral input. Please use the Legacy Exodus Reader.
• Fixed an error with build_cube() where corner nodes were scaled twice

[roystgnr]

I'd have split this into like 6 commits instead of 2 (re-bootstrap should be separate, tests could be separate, additional build_cube option should be separate), but the contents look mostly solid. That hardcoded block id (unless I'm just misunderstanding?) is the only thing that looks wrong.

[roystgnr]

Looks like you turned on some auto-code-reformatting option in your editor? That works within MOOSE, so long as you're pointing clang-format to the MOOSE config file and you're using the right clang-format version, but libMesh doesn't use the same (or any clang-format config-describable) formatting standards, and auto-reformatting utterly wrecked your ff02c5c commit. I can't find the needle of the actual changes in the 4000-line haystack of the reformatting.

As a general safety rule, never git add some/file or git commit -a without git diff to look at the changes first to make sure you're not committing anything you didn't mean to. Even in repos like MOOSE where CI will block formatting changes, it's still easy to accidentally commit diagnostic code, accidentally put multiple distinct changesets into a single commit, etc.

[roystgnr]

The elem quality updates should have gone in a separate PR, but fortunately I'm not seeing any problems with them, so we might as well get them in via this one.

Civet's not letting me activate the jobs here, though; hopefully it's an easy fix, but we'll have to wait on CI before merging.

[roystgnr]

-Wshadow errors ... but the second one actually does look like poor name choice overlap.

[roystgnr]

Huh. These remaining errors are real, and there's more than one variety of them. I can reproduce some kind of error; I'll try to look into it soon.