Do not include inner corner cells in FCI

[dschwoerer]

The inner corner values (i.e. values in the parallel slices, that are in the inner x-boundary) are normally not set, and should thus not be used.

I think this can also go to next, and then be merged into the 3D branch.

[ZedThree]

I'm not sure I see any difference, this change is functionally identical? Am I missing something?

[dschwoerer]

Sorry, yes, that is just a cleanup - I thought the continue was confusing, if the else branch is so short.

Probably should have gone into another commit ...

[github-actions]

clang-tidy review says "All clean, LGTM! 👍"

[ZedThree]

Fedora failing test is a Timer unit test:

The difference between Timer::getTime("ResetTimeLabelInScope test") and elapsed.count() is 0.0005812180000000001, which exceeds bout::testing::TimerTolerance, where
Timer::getTime("ResetTimeLabelInScope test") evaluates to 0.0016420510000000001,
elapsed.count() evaluates to 0.001060833, and
bout::testing::TimerTolerance evaluates to 0.00050000000000000001.

This is probably fine and just one of the difficulties of unit testing timers