refactor: Use enum class for MaterialUpdateStage

[paulgessinger]

This change is motivated by arithmetic between non-class enums and other types that are not the same enum being deprecated in C++20. This mainly affects NavigationDirection, but I took the chance to update the MaterialUpdateStage enum to make it an enum class.

[codecov]

Codecov Report

Merging #1205 (981ec79) into main (f7c32a8) will increase coverage by 0.00%.
The diff coverage is 44.00%.

@@           Coverage Diff           @@
##             main    #1205   +/-   ##
=======================================
  Coverage   47.85%   47.86%           
=======================================
  Files         371      372    +1     
  Lines       19413    19443   +30     
  Branches     9148     9151    +3     
=======================================
+ Hits         9291     9306   +15     
- Misses       3791     3806   +15     
  Partials     6331     6331           

Impacted Files	Coverage Δ
Core/src/Definitions/Common.cpp	0.00% <0.00%> (ø)
...de/Acts/TrackFinding/CombinatorialKalmanFilter.hpp	33.24% <37.50%> (+0.26%)	⬆️
Core/include/Acts/TrackFitting/KalmanFitter.hpp	44.48% <50.00%> (+0.91%)	⬆️
Core/include/Acts/TrackFitting/detail/GsfActor.hpp	42.02% <57.14%> (+0.68%)	⬆️
Core/include/Acts/Material/ISurfaceMaterial.hpp	70.83% <100.00%> (+7.67%)	⬆️
...Propagator/detail/PointwiseMaterialInteraction.hpp	96.87% <100.00%> (ø)

📣 Codecov can now indicate which changes are the most critical in Pull Requests. Learn more

[stephenswat]

This one looks good to go as it - there is one thing I would change, but it's more of an opinion matter, so I'll stamp it and then you can decide where to go with it. 👍

[stephenswat]

Fantastic, thanks!

[paulgessinger]

Updated after GSF went in, @stephenswat can you reapprove?

[paulgessinger]

Actually, I now also removed the arithmetic operator overload, since that's not needed anymore after the change requested by @stephenswat.

[stephenswat]