GitLab

#1509 Remove Docker generation, as it is handled by a brand new sibling project. A line has been added in the README.

#1507 Keep the tuple of variadic arguments fed to the SetComputeEltArrayArguments as a tuple of strong types to have the proper one selected in the overload resolution for the SecondPiolaKirchhoffOperator.

#1507 Implement strong types for every Assemble overload of the SecondPiolaKirchhoffOperator. Modified temporarly yml configuration for the CI build.

#1507 Implement strong type for the interface of the first Assemble overload of the SecondPiolaKirchhoffOperator.

#1505 Refactoring of VariationalFormulation concerning non linear solve: the upgrade to PETSc 3.12 underlined several issues with the way I was using it, and they made a change that made hyperelastic model be way less efficient due to a poor line search choice. So now:

- SnesInterface is always used for the function and the jacobian function used in SNES.
- There are no longer VariationalFormulation::ImplementSnesFunction() to fill in derived variational formulation; the user must now just define a ComputeResidual() and ComputeTangent() with arguments.
- And this is only if new VariationalFormulation template argument indicates non linear solve is enabled; if not it is not necessary to put anything in the derived model (in contrast to what was done previously: ImplementSnesFunction was defined and returned nullptr for instance).
- Line search type may be modified; by default _basic_ is used explicitly (thus avoiding PETSc using another one without the developer knowledge - I spent countless hours trying to understand why the convergence suddenly was less efficient)

#1504 Replace entirely boost Filesystem by std::filesystem. It is currently done by replacing the call in the File.cpp and Directory.cpp files; the modifications have been kept as minimal as possible from previous version with Boost (I will just remove the intermediate catch blocks in next commit as contrary to Boost exception the exception involved here derives from std::exception and is thus catchable directly in my main programs).

It is more than likely that I could write a more fluent code to handle the filesystem using the new C++ interface (which I do not master yet) but in the occurrence I have chosen to apply the "If it ain't broken don't fix it" principle.

#0 Fix warnings that appeared following the update of third party libraries (to v20.04 of ThirdPartyCompilationFactory).

#1502 Modified the ensight output so that the header of result files for each unknown is related to its nature (scalar or vectorial). Modified expected results accordingly.

#1444 More Ensight6 refactoring... and yet not done. In previous version I in fact wrote several times the same file... A proof that streamlining the code to better understand how it works is necessary and will be appreciated the day EnsightOutput will be revisited next (or as an inspiration when another output format is to be written).

#0 Improve in tests about Ensight results the message to specify exactly the paths of the files being compared in the output message.

#1444 Rewriting of Ensight output in progress; does not work yet. At the moment, the IsRefinedMesh argument has been ignored; I'll see if it needs to be handled separately or not (to be frank I'm not sure the feature still works properly: it's not in the tests and isn't used by anyone to my knowledge).

#1443 Modify slightly API of a method: ComputeNumberingSubsetList() now returns the reference directly, avoiding to add a line to do this step.

#859 XCode: fix a file poorly attributed (GracefulExit was unduly in Core) and remove entirely a directory that caused a crash under macOS Catalina.