Pde remove jk

[joergfunger]

This is a first try to get this remove jk stuff running. A meaningful benchmark is however not possible. This is due to several issues that have to be further addressed.

• There are many conversions between Eigen and DofVectors. This is true for the Hessian/Gradient, but also for the Cmat
• One of the reasons is that we require DofVectors as Input/Output of the Newton scheme. However, internally we solve with eigen matrices. This has some weird effects, such as the norm of the residual as a dofVector not being zero (since the constraint dofs are included), but when converting it, those components are removed. From my point of view, this nonzero residual vector for (real) vanishing residual seems counterintuitive.

[codecov[bot]]

Codecov Report

Merging #238 into PDE_reviewed will decrease coverage by 0.02%. The diff coverage is 98.02%.

@@               Coverage Diff                @@
##           PDE_reviewed     #238      +/-   ##
================================================
- Coverage         84.45%   84.42%   -0.03%     
================================================
  Files               296      294       -2     
  Lines             10754    10774      +20     
================================================
+ Hits               9082     9096      +14     
- Misses             1672     1678       +6

Flag	Coverage Δ
#integrationtests	62.21% <95.81%> (-0.13%)	:arrow_down:
#unittests	87.71% <88.88%> (-0.21%)	:arrow_down:

Impacted Files	Coverage Δ
nuto/mechanics/constraints/Constraints.h	100% <ø> (ø)	:arrow_up:
nuto/mechanics/tools/NodalValueMerger.h	100% <ø> (ø)	:arrow_up:
nuto/mechanics/cell/SimpleAssembler.h	100% <ø> (ø)	:arrow_up:
nuto/mechanics/tools/QuasistaticSolver.h	100% <ø> (ø)	:arrow_up:
nuto/mechanics/tools/TimeDependentProblem.h	100% <ø> (ø)	:arrow_up:
nuto/mechanics/tools/TimeDependentProblem.cpp	97.56% <100%> (-0.06%)	:arrow_down:
test/mechanics/cell/SimpleAssembler.cpp	100% <100%> (ø)	:arrow_up:
nuto/mechanics/cell/SimpleAssembler.cpp	100% <100%> (ø)	:arrow_up:
test/mechanics/dofs/DofVector.cpp	100% <100%> (ø)	:arrow_up:
nuto/mechanics/tools/NodalValueMerger.cpp	100% <100%> (ø)	:arrow_up:
... and 11 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 517ceb7...3bdd4b2. Read the comment docs.