Improve dof distribution interface

[termi-official]

This enables e.g. order 3 Serendipity and arbitrary order Lagrange in 2D with the MixedDofHandler. Closes #661.

TODOs

• Basic fix to distribute high order vertex dofs and high order edge dofs
• Some test coverage
• Fix mixed tests (faulty semantics?)
• Devdocs

[codecov-commenter]

Codecov Report

Patch coverage: 88.88% and project coverage change: -0.11 ⚠️

Comparison is base (f923711) 92.90% compared to head (f78f884) 92.79%.

📣 This organization is not using Codecov’s GitHub App Integration. We recommend you install it so Codecov can continue to function properly for your repositories. Learn more

Additional details and impacted files

@@            Coverage Diff             @@
##           master     #627      +/-   ##
==========================================
- Coverage   92.90%   92.79%   -0.11%     
==========================================
  Files          30       30              
  Lines        4324     4343      +19     
==========================================
+ Hits         4017     4030      +13     
- Misses        307      313       +6     

Impacted Files	Coverage Δ
src/Dofs/DofHandler.jl	92.60% <87.50%> (-1.66%)	⬇️
src/Dofs/ConstraintHandler.jl	95.95% <100.00%> (ø)
src/Grid/grid.jl	91.62% <100.00%> (+0.04%)	⬆️
src/interpolations.jl	98.13% <100.00%> (+<0.01%)	⬆️

☔ View full report in Codecov by Sentry.
📢 Do you have feedback about the report comment? Let us know in this issue.

[termi-official]

Foundation for the vector wrapper discussed in #676 . I will introduce the interface for the vector stuff in the next PRs. I am still playing around.

[termi-official]

Now, this should contain everything for #676 . I will fix the remaining allocation issue if the interface is okay.

An open issue is to fix the dics used during distribution, as the dof distrubtion currently only works by chance for the implemented interpolations.

[termi-official]

This should be almost there. However, there is quite a performance regression (~5%) over master on close! performance, although the total number of allocation is reduced. I tried to track this one down for several hours but could not find why exactly. :/