Port elastic-tube-1d solvers to version 3

[fsimonis]

This PR ports the elastic-tube-1d cpp solvers to version 3

This also rearranges the solvers to follow the pattern read data -> solve -> write data -> advance.

Also fixes the incorrect coordinates in solid-cpp #340

[fsimonis]

Closes #395

[valentin-seitz]

I just started a small system test job: https://github.com/precice/tutorials/actions/runs/7492877499/job/20397361340
And apparently, field-compare reports some major differences between the v2.5 reference data (we use the exchanged data between the participants) and the current status. (see stdout.log in the respective folders)
I just wanted to share this. It could also be a bug in the tests or some v2 vs. v3 hiccup.

[IshaanDesai]

I ran several combinations of the C++ and Python participants and everything seems to be working.

[fsimonis]

@valentin-seitz The solid-cpp solver was bugged and produced very bad results before 7de3813.
Could that be it?

Solid Cpp	Solid Python

[fsimonis]

@valentin-seitz I checked the uploaded run of the workflow you linked to.

For the solid-python fluid-python case, fieldcompare fails, but the exports are close to identical apart from some minor differences. I checked provided meshes only. Is there a way to get more detailed information on what differs?

[fsimonis]

I ran the version 2 of the elastic-tube-1d with the watchpoint at x=5 on the Fluid Solver and then plotted them using the state of #438 (this PR + #435 + watchpoints and plotting).

The result is this:

The results of the v2 python solid match the results of this PR.
The buggy (too coarse) solid solver of v2 leads to different results (smaller max diameter in oscillations)

[fsimonis]

I'll merge this. We can investigate possible regressions later.