Port most of the compass framework

[xylar]

This merge brings over a good portion of the compass framework. It also:

• renames compass to polaris nearly everywhere (but leaves encompass rather than changing it to enpolaris ;-))
• renames "MPAS cores" to "components" in the code and documentation
• refers to "E3SM components" rather than "MPAS models"
• changes compass run to polaris serial in anticipation of polaris run being the default, task-parallel approach in the near future
• Updates to mpas_tools 0.18.0 to fix an issue with reading SciColorVis colormaps
• Adds a bare-bones ocean component with the cosine bell test case
• Adds a chunk of documentation for both the User's and Developer's Guides in MyST format

Checklist

• [x] User's Guide has been updated
• [x] Developer's Guide has been updated
• [x] Documentation has been built locally and changes look as expected
• [x] Testing comment in the PR documents testing used to verify the changes
• [x] New tests have been added to a test suite

[xylar]

Testing

I have build the documentation locally. While there are many complaints about missing sphinx redirects (because of missing documentation yet to be ported), it builds and has no other errors I noticed.

I have built conda and spack environments on Chrysalis using ./configure_polaris_env.py and have used it to successfully run both cosine bell test cases (QU and Icos).

Here is the configure command I used:

SCRATCH=/lcrc/group/e3sm/ac.xylar
./configure_polaris_envs.py \
    --conda ~/chrysalis/mambaforge/  \
    --env_name polaris_test  \
    --update_spack  \
    --mache_fork xylar/mache  \
    --mache_branch fix_chrysalis_os  \
    --recreate  \
    --spack $SCRATCH/spack_test  \
    --tmpdir $SCRATCH/spack_temp \

[xylar]

@cbegeman, this is a monstrosity . Would you be willing to review it in some capacity? Perhaps a minimum might be to run the configure script, similarly to what I did above (building the spack env takes many hours), build the docs locally, and see if you can also run the cosine bell. Anything more substantial would be up to you. There are bound to be errors in both the docs and the code port but it is so many files (71 currently) that you can't hope to look over even a small fraction of them.

[cbegeman]

@xylar I'd be happy to review!

[xylar]

The mache branch I was testing with got merged, so now the instructions would be:

SCRATCH=/lcrc/group/e3sm/${USER}
./configure_polaris_envs.py \
    --conda ~/mambaforge/  \
    --env_name polaris_test  \
    --update_spack  \
    --mache_fork E3SM-Project/mache  \
    --mache_branch main  \
    --recreate  \
    --spack $SCRATCH/spack_test  \
    --tmpdir $SCRATCH/spack_temp

[cbegeman]

@xylar I have pushed a few commits that make the cosine bell case a bit more suitable as an "ideal" test case, as we discussed. Please edit or request edits as needed. I confirmed that the diffs introduced by the initial state for the cosine bell case are similar to those in https://github.com/MPAS-Dev/compass/pull/519.

[xylar]

@cbegeman, your additions look great to me! I just want to run the tests one last time to do my due diligence.

[xylar]

I reran my tests on Chrysalis with @cbegeman's 4 new commits and everything looks great!