Restructure runscripts and related resources

[MakisH]

Similarly to #169 and #166, this collects all the run scripts into common files, to reduce duplication.

• OpenFOAM: separated between a "preparation" and an "execution" phase. The execution is same for every case, the preparation can differ. I did not find any reason to use a function, so I just call the general script from inside each case-specific script.
  • I also took the opportunity to clean up some old crimes in the heat exchanger, so this also closes heat-exchanger OpenFOAM is missing run.sh #171.
  • Since this case is special because we download some additional files, I added these extra files in smaller .gitignore files, so this also rouches aspects of Inconsistent .gitignore among cases #87.
• deal.II: no changes.
• CalculiX: I only removed unnecesary/inconsistent comments and variables. Some cases with large meshes already set (default) values for OpenMP, I assume that these actually work and can be useful.
• SU2: We only have one case, I left it hard-coded and simplified it.
• FEniCS: In some cases, I added a run.sh or a clean.sh that was missing. I guess that some of the cases are not yet in the fully restructured form, but these are elements we need anyway. I have not checked if the runscripts run. I also softened a bit an existing .gitignore file for a FEniCS case.
• Nutils: Similarly to FEniCS. I also extended the clean_nutils to remove all ./*.vtk files. Not tested.
• code_aster: Only formatting and consistency changes.

All scripts have been validated with shellcheck.

Before merging:

• Check documentation for renamed scripts: plot displacements, download meshes, run.sh / runFluid / Allrun, ...
• Check documentation for 0.orig (moved to 0 in most cases) . What happens when we use the cleaning scripts of the foundation version?

After merging:

• Update Quickstart (@davidscn would be a good review/cross-check that you apply the same changes there and I review the separate PR)

[davidscn]

Very nice! I guess your checklist keeps track of already checked run scripts and we can check tomorrow if everything works.

[MakisH]

@davidscn ready for review! If you can easily run something with Nutils/FEniCS, please do. Other than than, any comments are welcome. Especially: did I forget anything? Is there anything in the documentation to fix?

[MakisH]

This dummy is generated at least in the heat exchanger CalculiX case, I have no idea where it comes from.

[MakisH]

Now we delete all the VTK files inside Nutils cases. I guess not much can go wrong.

[davidscn]

A two more comments:

• the OpenFOAM run ends for me with the following message ../../tools/run-openfoam.sh: 17: .: openfoam-remove-empty-dirs.sh: not found
• I created the deal.II run script as symbolic link similar to our cleaning scripts. Do you want to keep it in case of OpenFOAM as it currently is?

I will adjust the quickstart in #145 and we can check the remaining parts during the coding days :)

[MakisH]

* the OpenFOAM run ends for me with the following message `../../tools/run-openfoam.sh: 17: .: openfoam-remove-empty-dirs.sh: not found `

Good catch, I fixed it by moving the cleaning part into run.sh as a separate step.

* I created the deal.II run script as symbolic link similar to our cleaning scripts. Do you want to keep it in case of OpenFOAM as it currently is?

In the way I see it, we cannot do this for OpenFOAM: some cases have different preparation steps and for all of them I create a named .foam file (opening two .foam files in ParaView should keep it clear what is what).

I will adjust the quickstart in #145 and we can check the remaining parts during the coding days :)

Then I don't wait for the Quickstart :).

[MakisH]

This PR is ready from my side. I also looked at the diff and it is as intended. Are there further comments or things to check?