CylBinaryCompactObject: add equiangular map.

[markscheel]

Proposed changes

Adds an equiangular map option to CylindricalBinaryCompactObject.

Upgrade instructions

When using the CylindricalBinaryCompactObject domain creator one must now specify UseEquiangularMap. Setting this option to false gives the previous behavior.

Code review checklist

• The code is documented and the documentation renders correctly. Run
  make doc to generate the documentation locally into BUILD_DIR/docs/html.
  Then open index.html.
• The code follows the stylistic and code quality guidelines listed in the
  code review guide.
• The PR lists upgrade instructions and is labeled bugfix or
  new feature if appropriate.

[nilsvu]

Are you relatively certain that the unit tests would catch bugs where coord maps join discontinuously because of missing equiangular maps? For example, do they fail when you only make some of the cylinders equiangular?

[markscheel]

I tried making only some of the wedges equiangular, or some of the cylinders equiangular, and the test still passes. I don't understand why TestHelpers::domain::creators::test_domain_creator doesn't test this. What happens for the other BBH domain? Is that untested as well?

Made other changes except for cartesian_product.

[kidder]

I tried making only some of the wedges equiangular, or some of the cylinders equiangular, and the test still passes. I don't understand why TestHelpers::domain::creators::test_domain_creator doesn't test this. What happens for the other BBH domain? Is that untested as well?

Made other changes except for cartesian_product.

try cherry-picking the commit from https://github.com/kidder/spectre/tree/check_block_faces and see if the test still passes

[markscheel]

Rebased, which also did some of the requested changes here. So now there are only 3 relevant fixup commits.
I will try https://github.com/kidder/spectre/tree/check_block_faces

[markscheel]

Thanks @kidder ! when I cherry-pick https://github.com/kidder/spectre/tree/check_block_faces, the current PR passes the tests as-is, but if I intentionally introduce a bug by turning on use_equiangular_map for some blocks but not others, then the test fails.

[kidder]

okay, I'll clean that up and make a PR

[nilsvu]

I thought that continuity of maps was tested in one of the generic tests already, but apparently it isn't. Thanks for checking. We don't need to hold up the PR with that. @kidder thanks for adding that test.

[kidder]

the current test only checks the corners of the block faces, not the interior...

[markscheel]

squashed

[nilsvu]

The test timed out in one of the builds. I've seen timeouts in this test before. Can we reduce the amount of work done in the big cartesian product of options?
Edit: or we randomly skip some of the iterations of the loop ("fuzzy cartesian product")

[nilsvu]

@markscheel would you mind adding the random-skipping code also to the BinaryCompactObject domain test while you rebase/squash? It's been timing out as well.