Make GSets better inferreable

[lgoettgens]

By adding a type parameter which represents the element type of the G-set (derived from the eltype of the seed iterator).
This then allows to concretize the return types of orbits and elements to something concretely that is statically known, i.e. in these two functions there is no dynamic dispatch happening, and they infer their concrete return type.

I noticed this issue while profiling some new functionality in #4053, since half of the flame graph was filled with dynamic dispatch stuff.

[ThomasBreuer]

Thanks.
Just for curiosity: If we introduce a type parameter that describes the elements of the G-set then wouldn't it be more logical to take the element type itself and not the type of the seeds?

[lgoettgens]

Thanks. Just for curiosity: If we introduce a type parameter that describes the elements of the G-set then wouldn't it be more logical to take the element type itself and not the type of the seeds?

That is indeed a good question. I'll play around with it tomorrow to see if I can get it working like that

[ThomasBreuer]

Looks good. Thanks.

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 84.63%. Comparing base (5edd2b0) to head (c675cd4).
Report is 6 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4052      +/-   ##
==========================================
+ Coverage   84.62%   84.63%   +0.01%     
==========================================
  Files         600      600              
  Lines       82590    82610      +20     
==========================================
+ Hits        69894    69921      +27     
+ Misses      12696    12689       -7     

Files with missing lines	Coverage Δ
src/Groups/group_characters.jl	95.65% <100.00%> (+0.01%)	⬆️
src/Groups/gsets.jl	91.35% <100.00%> (-0.04%)	⬇️

... and 6 files with indirect coverage changes