A postprocessor that outputs the second invariant of the deviatoric stress tensor

[anne-glerum]

For all pull requests:

• I have followed the instructions for indenting my code.

For new features/models or changes of existing features:

• I have tested my new feature locally to ensure it is correct.
• I have created a testcase for the new feature/benchmark in the tests/ directory.
• I have added a changelog entry in the doc/modules/changes directory that will inform other users of my change.

[bobmyhill]

Could the calculation of the stress tensor be moved into a utilities function? There's quite a lot of code overlap with the Stress visualization function (which is currently in flux; see #4375).

[anne-glerum]

I agree that it would make sense not to repeat assembling the stress tensor in different postprocessors. Currently there are 4 pps that could benefit from a utilities function:

stress.cc
shear_stress.cc
maximum_horizontal_compressive_stress.cc
principal_stress.cc

However, I think it falls outside the scope of this current PR to create such a function. Happy to open a new PR of course.

[gassmoeller]

This looks generally ok to me, but I have one question (see above). Ready if you fix this here, or explain why this PR does not need the fix.

[gassmoeller]

Ready to merge, but I forgot to ask for a changelog entry. Could you add one? Feel free to merge yourself afterwards (or leave me a note so I can merge).

[anne-glerum]

I'll add the changelog, but should I undo the change in the way the stress is computed when elasticity is used? Seeing that we decided to not change the other stress postprocessors, undoing the last commit would at least make all the stress postprocessors consistent.

[gassmoeller]

No don't undo the last commit. The current form is correct with the current stress calculation. The fact that it is inconsistent with the other postprocessors is caused by the other postprocessors being wrong at the moment. Let's not break stuff because we decided we will make changes in the future (let's just remember to change this postprocessor when we finally do).

To remember: Related to #4375, #4370

[anne-glerum]

Okay, I've added the changelog.