Add hydrometeors

[rtodling]

Description

This adds hydrometeors and changes to the linear obs operators necessary to account for sensitivity of all-sky MW to such fields.

Dependencies

• [https://github.com/JCSDA-internal/fv3-jedi/pull/1257 ] waiting on fv3-jedi
• [ https://github.com/JCSDA-internal/saber/pull/939] waiting on saber

Impact

Enhanced features and corrections and consistency w/ GSI

Issues

[#418] this is associated w/ the issue marked here.

[rtodling]

Look around to try to understand some - unrelated to this - I came across the fact that by default, the hydrometeors radii is set to use GFDL's numbers ... the GSI numbers are an option ... The yaml's for GMI and AMRS2 in swell are using the fv3-jedi default which is not what we want - I believe.

I ran a little test w/ GMI obs only to see what difference we get from these two schemes in comparison to GSI: I find that the difference between the two schemes is considerable smaller than the difference of either scheme w/ the actual GSI. Here is an illustration of the differences: top: GSI-minus-fv3-jedi using GSI radii; bottom: diff of JEDI increments when using either GSI or GFDF radii ...

The differences in the temperature are unnoticeable either way - GSI and JEDI are quite good match to each other regardless of which scheme is used - no need to show.

In any case, the consistent choice of radii is also being introduced in the changes presented here.

[rtodling]

The PRs from JEDI mentioned above have been merged. I believe the changes here could now make it into SWELL

[Dooruk]

@rtodling Can you set the environment LOG_INFO variable to 1 and run swell test code_tests with this branch please? There is a code testing issue.

https://geos-esm.github.io/swell/#/code_tests/code_tests

[Dooruk]

This will be on-hold until a JEDI build post ~September 20th.