Add save_MOM6_internal_state subroutine to MOM.F90 and call during restart

[cspencerjones]

As described in issue #372, I would like to be able to create restart files that contain information about the particle location. These files should be written at the same time as other restart files. I cannot add these calls directly to the driver, because the driver does not have information about the particle location.

We have opted to add save_MOM6_internal_state as a subroutine in MOM.F90, and to call it from each of the drivers. We hope this will allow for more new packages to write restart files in the future.

I have tested these additions for solo_driver and FMS_cap, but I don't have access to the other drivers, so I haven't been able to test them.

cc @adcroft

[cspencerjones]

Since Alistair is one of the authors of this PR, it probably doesn't make sense for him to be a reviewer. Can you assign someone else @marshallward ?

[marshallward]

I can review it (hopefully very soon).

[codecov]

Codecov Report

Merging #375 (70f33aa) into dev/gfdl (77b5881) will decrease coverage by 0.01%.
The diff coverage is 50.00%.

❗ Current head 70f33aa differs from pull request most recent head a7e4a6f. Consider uploading reports for the commit a7e4a6f to get more accurate results

@@             Coverage Diff              @@
##           dev/gfdl     #375      +/-   ##
============================================
- Coverage     38.22%   38.22%   -0.01%     
============================================
  Files           269      269              
  Lines         76413    76417       +4     
  Branches      14031    14030       -1     
============================================
- Hits          29209    29208       -1     
- Misses        41962    41967       +5     
  Partials       5242     5242              

Impacted Files	Coverage Δ
config_src/drivers/solo_driver/MOM_driver.F90	50.82% <33.33%> (-0.22%)	⬇️
src/core/MOM.F90	51.45% <66.66%> (+0.09%)	⬆️

... and 1 file with indirect coverage changes

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[Hallberg-NOAA]

This seems to me like a reasonable solution to the problem with calling the routines to save the restart files for submodules that are known to MOM6 but not to the driver, as discussed in the description of this PR. As noted in the comments in this commit, there is more work to be done along these lines, but this is a viable way-point in the code changes. This can be merged into dev/gfdl in due course, once the pipeline regression testing has passed.

[marshallward]

Given that this changes the drivers, which needs consultation with the partners, should we just push ahead with the suggested TODO plan and introduce save_MOM_restart which calls save_restart?

I can work on this and submit it as a PR to this PR.

[cspencerjones]

I'm happy for you to do that @marshallward , as long as it doesn't take months and months.

[marshallward]

I've submitted a potential revision to this PR which does the following:

• merges save_restart and save_MOM6_internal_state into a generalized restart function.

• Removes MOM_restart dependencies from the primary model driver code. (It is still present in the surface forcing modules).

The PR is here for review: cspencerjones#1

It's potentially disruptive, so I expect it needs careful review before merging. If it's too much, then maybe we can leave it for a later date.

[marshallward]

We can take this as-is, and the suggested amendments will be handled in a separate PR.

[marshallward]

Gaea regression: https://gitlab.gfdl.noaa.gov/ogrp/MOM6/-/pipelines/19805 ✔️

[cspencerjones]

Thanks very much @marshallward !!