write a module describing orientation / general orientatation cleanup

[drwells]

We should put this information all in one place. A few relevant classes/functions/etc:

1. TriaObjectsOrientations
2. ReferenceCell
3. GeometryInfo
4. the simplex module
5. make_periodicity_constraints
6. QProjector (a lot of orientation information is hard-coded into this class)
7. GridTools::orthogonal_equality (this one uses std::bitset<3>)
8. PeriodicFacePair (also uses bitset)

In a similar vein, we should settle on 'combined orientation' or 'raw orientation' - I recall seeing both in the library

Consistency problems:

• combined vs raw orientation (we should settle on a single name)
• different conventions for converting (bool, bool, bool) to a single integer flag (e.g., in FiniteElement)
• MatrixFree wants the standard orientation to be all zeros, whereas 1u is the default orientation in most of the rest of the library
• we should not use both unsigned char and std::bitset<3>
• Standardize on using face_orientation as the sole relevant bit in 2D. Many parts of the library (e.g., GeometryInfo) use face_flip instead.
• Standardize on using the order 'orientation, rotation, flip'. The std::bitset<3> version uses 'orientation, flip, rotation'.
• Add utility functions for getting reverse orientations instead of hard-coding it
• We sometimes assume that things are always in the standard orientation in 2d. We should remove this assumption because it doesn't work for mixed meshes (i.e., FE_Q cannot make this assumption since it does not know whether or not it is part of a mixed mesh).
• Introduce types::orientation_type or something similar instead of using unsigned char
• Clean up the documentation of QProjector. It was written long ago and the style is difficult to understand (it focuses too much on how and not enough on why)
• FE_ABF doesn't actually loop over the different possible face orientations - it used (false, false, false) (which is almost surely wrong) to initialize some values. Take a closer look at how it computes things.
• Take another look at face_to_cell_index(): maybe we have enough infrastructure now to generalize everything.

[bangerth]

It should be in ReferenceCell. That's where we've been moving all of the other stuff that relates to geometric descriptions of reference cells.

[drwells]

See also #15025.