#1664 TyingPointDataForComponent objects now compute the relevant local quantities at quadrature points when there is no specific interpolation at tying points. Greatly simplifies the API in the NonlinearShell operator (at the cost of some unnecessary copies during initialization).

#1664 Shell operator now correctly handles cases where the test function is not the same as the displacement one. Some further refactoring can be done for the TyingPointDataForComponent class so that it also computes the values at the quadrature points directly when there is no specific interpolation for a component (at the time it is the NonlinearShellOperator itself that does the computation through constexpr evaluations).

#1660 Tying point data now treated as a policy. Still have to deal with the cases where not every component of the strain tensor are interpolated (for MITC9 Q1P2 for instance) through constexpr evaluations.

#1660 Refactored tying point data in a standalone object, which will be used as a policy for the Shell operator in order to choose an interpolation rule.

#1664 Shell operator implemented for Q2 elements, after a long due rebase on master. Tying points policy needs to be added and documentation remains to be improved.

#0 Implemented some temporary outputs in the ModelInstances for the validation of the Shell operator.

#1586 Implemented tying point interpolation, Shell operator now functional but remains to be validated. Added a ModelInstance with shell elements and a sequential post processing to generate a Q2 visualization compatible with ViZiR.

#1586 Implemented tying points data (storage/API needs improvement). Coords no longer have an Interface type (some testing is required to make sure that relevant geometric quantities can still be computed correctly.)

#1586 Removed unnecessary template parameters for the hyperelasticity policy so that we only define them at the hyerelastic law level.

#1586 Moved some of the heavy computation at the Constructor level so that we only do it once and updated documentation.

#1586 Rewritten every contribution (residual/tangent) so that we can directly use reinterpolated values. Some refactoring remains to be done and some of the computations should only be done during the initialization phase.

#1586 Hyperelasticity in generalized coordinates now working. I4 is yet to be implemented (need to compute the contravariant basis). Documentation remains to be improved, as well as some refactoring of the implementation. Optimization could be done later on as well.

#1586 Computation of local matrices needed by the the Shell operator. Made one of the hyperelastic law template so that the proper invariant holder type is used (will have to be applied to all the other hyperelastic laws).

#1586 Implemented 3D metric tensors. Displacement gradient as well as the Cauchy-Green tensor are computed in generalized coordinates.