Ensure Time dim is added to landice initial condition

[xylar]

MPAS meshes include an empty, unlimited Time dimension but the latest NetCDF4 drops this dimension since no variable use it. The create_from_generic_mpas_grid() funciton in landice.create was taking advantage of Time being a dimension in the MPAS mesh but this should no longer be assumed. This PR fixes that assumption by explicitly adding Time to the list of dimensions to create if it is not already in the input dataset.

[xylar]

@matthewhoffman and @trhille, a new version of JIGSAW was released last week:
dengwirda/jigsaw-python#29

In the process of testing it, I learned that new newest version of xarray breaks several of our workflows. I reported this to them:
pydata/xarray#10647
it has been fixed in:
pydata/xarray#10648
and I added a workaround for us in:
#662
and, after performing Polaris, MPAS-Ocean and MPAS-Analysis testing, I made a new release of MPAS-Tools with the fix:
https://github.com/MPAS-Dev/MPAS-Tools/releases/tag/1.3.0

I didn't do any MALI testing that that has proven to be a dumb move. MALI breaks for a different reason. There has also been an update to NetCDF4:
https://github.com/Unidata/netcdf4-python/releases/tag/v1.7.2rel
I'm not sure if the change was in this release or a previous one but I suspect the most recent release. In any case, it seems that Time is now being dropped when NetCDF4 reads in an MPAS mesh dataset as described above (because it is not used in any variables). As a result, we need to add it explicitly or we get errors like:

Traceback (most recent call last):
  File "/gpfs/fs1/home/ac.xylar/chrysalis/miniforge3/envs/dev_compass_1.8.0-alpha.2/bin/create_landice_grid_from_generic_mpas_grid", line 9, in <module>
    sys.exit(create_from_generic_mpas_grid())
             ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
  File "/gpfs/fs1/home/ac.xylar/chrysalis/miniforge3/envs/dev_compass_1.8.0-alpha.2/lib/python3.13/site-packages/mpas_tools/landice/create.py", line 284, in create_from_generic_mpas_grid
    newvar = fileout.createVariable('thickness', datatype, ('Time', 'nCells'))
  File "src/netCDF4/_netCDF4.pyx", line 3005, in genexpr
  File "src/netCDF4/_netCDF4.pyx", line 3005, in genexpr
  File "/gpfs/fs1/home/ac.xylar/chrysalis/miniforge3/envs/dev_compass_1.8.0-alpha.2/lib/python3.13/site-packages/netCDF4/utils.py", line 45, in _find_dim
    raise ValueError("cannot find dimension %s in this group or parent groups" % dimname)
ValueError: cannot find dimension Time in this group or parent groups

      Failed
Exception raised while running the steps of the test case
Traceback (most recent call last):
  File "/gpfs/fs1/home/ac.xylar/compass/update-to-mpas-tools-1.3.0/compass/run/serial.py", line 322, in _log_and_run_test
    _run_test(test_case, available_resources)
    ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/gpfs/fs1/home/ac.xylar/compass/update-to-mpas-tools-1.3.0/compass/run/serial.py", line 419, in _run_test
    _run_step(test_case, step, test_case.new_step_log_file,
    ~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
              available_resources)
              ^^^^^^^^^^^^^^^^^^^^
  File "/gpfs/fs1/home/ac.xylar/compass/update-to-mpas-tools-1.3.0/compass/run/serial.py", line 470, in _run_step
    step.run()
    ~~~~~~~~^^
  File "/gpfs/fs1/home/ac.xylar/compass/update-to-mpas-tools-1.3.0/compass/landice/tests/dome/setup_mesh.py", line 78, in run
    check_call(args, logger)
    ~~~~~~~~~~^^^^^^^^^^^^^^
  File "/gpfs/fs1/home/ac.xylar/chrysalis/miniforge3/envs/dev_compass_1.8.0-alpha.2/lib/python3.13/site-packages/mpas_tools/logging.py", line 66, in check_call
    raise subprocess.CalledProcessError(process.returncode, print_args)

Note: MPAS-Tools lints and formats any files that we change using ruff-format (a standard linting tool also used in Polaris and across the E3SM ecosystem of python packages). I have updated landice.create to conform to this standard in a separate commit. Please ignore that commit in your review unless you are interested in the linting.

[matthewhoffman]

@xylar , thank you for taking care of this. I imagine it was a lot of work to untangle the new issues with xarray and netcdf4. I reviewed the first commit by inspection and it makes sense to me.