Removed some code duplication

[fweik]

by making some utils math functions usable on the gpu.
Also replaced the macro constants by functions, also for gpu.

[codecov]

Codecov Report

Merging #2697 into python will increase coverage by <1%.
The diff coverage is 92%.

@@           Coverage Diff           @@
##           python   #2697    +/-   ##
=======================================
+ Coverage      79%     79%   +<1%     
=======================================
  Files         506     508     +2     
  Lines       27038   27050    +12     
=======================================
+ Hits        21484   21496    +12     
  Misses       5554    5554

Impacted Files	Coverage Δ
src/core/utils/Span.hpp	100% <ø> (ø)	⬆️
...rc/core/utils/interpolation/detail/ll_and_dist.hpp	100% <ø> (ø)	⬆️
src/core/reaction_ensemble.cpp	81% <0%> (ø)	⬆️
src/core/immersed_boundary/ibm_tribend.cpp	80% <0%> (ø)	⬆️
src/core/nonbonded_interactions/ljcos.cpp	100% <100%> (ø)	⬆️
.../core/electrostatics_magnetostatics/p3m-common.cpp	62% <100%> (ø)	⬆️
.../electrostatics_magnetostatics/mdlc_correction.cpp	88% <100%> (ø)	⬆️
src/core/electrostatics_magnetostatics/p3m.cpp	83% <100%> (ø)	⬆️
...core/electrostatics_magnetostatics/p3m-dipolar.hpp	100% <100%> (ø)	⬆️
src/core/bonded_interactions/bonded_tab.cpp	93% <100%> (ø)	⬆️
... and 15 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 5055147...00377f4. Read the comment docs.

[fweik]

This does not work with HIP, but the error message is not really helpful...

[fweik]

@KaiSzuttor the HIP stuff could be fixed, I also replaced the constant macros that collided with walberla (and should have been replaced anyway).