Not checkpointing non-geometric fields

[vidulejs]

Note: This Pull Request has been superseded and is now closed.
The work originally done in this PR has been cleaned up and split into two new PRs which solve two different issues.

• Mesh points & volumes checkpointing #369
• FSI implicit subcycling fix #370

Main changes

• Fixed meshOldPoints checkpointing. Previously mesh would move from incorrect interim state to meshPoints, instead of meshOldPoints (leads to incorrect mesh flux).
• Removed comments and vectors to mesh fields which don't exist
• Mesh volume checkpointing (needed for higher order time derivative schemes)

[precice-bot]

This pull request has been mentioned on preCICE Forum on Discourse. There might be relevant details there:

https://precice.discourse.group/t/coupling-problem-about-in-house-openfoam-solver-with-calculix-by-precice/1125/6

[vidulejs]

Closing this draft pull request in favor of two cleaner PRs.

This branch became a long-running experiment for both mesh checkpointing and an
FSI subcycling bugfix. To make the review process clearer and more manageable, I
have split the work into two separate, clean pull requests:

1. Mesh points & volumes checkpointing #369: This PR contains the complete,
   consolidated history for the mesh and volume checkpointing feature.
2. FSI implicit subcycling fix #370: This PR builds on the checkpointing branch
   to fix a bug in how readCouplingData handles time during implicit subcycling steps.