Bug fix to the HAILCAST diagnostic code

[jaymes-kenyon]

Description

This PR fixes a units bug in the HAILCAST diagnostic code.

An identical PR was merged into the dev/emc branch of NOAA-GFDL/GFDL_atmos_cubed_sphere; see NOAA-GFDL#320 . This PR accomplishes the same fix to the production/RRFS.v1 branch of the NOAA-EMC fork.

How Has This Been Tested?

Testing details are available via the previous PR at the link above.

Checklist:

Please check all whether they apply or not

• [ x] My code follows the style guidelines of this project
• [ x] I have performed a self-review of my own code
• [ x] I have commented my code, particularly in hard-to-understand areas
• [ x] I have made corresponding changes to the documentation
• [ x] My changes generate no new warnings
• [ x] Any dependent changes have been merged and published in downstream modules

[jaymes-kenyon]

@MatthewPyle-NOAA — Wanted to bring this PR to your attention, given the pending RRFSv1 science code freeze. Thanks!

[MatthewPyle-NOAA]

Sorry I hadn't seen this one...approving.

[MatthewPyle-NOAA]

@jaymes-kenyon Now we need to do PRs at the UFS WM level and FV3ATM levels to fully pull this fix into the RRFSv1 branch. Can you open those, or should I? Thanks!

[jaymes-kenyon]

@MatthewPyle-NOAA — I can probably tackle the follow-on PRs. A basic question: in the "fv3atm" repo, what file / component needs to be modified in order to update the hash that "atmos_cubed_sphere" points to? The hash after the "@" sign (see below) is one of the needed updates, as I understand it (analogous to Externals.cfg). Thanks for the advice!

[MatthewPyle-NOAA]

Actually @jaymes-kenyon - it actually might not be needed. It seems like it was actually handled by another PR (fab198c#diff-6e6b355f9e32fd586e414c884cd7faa0f673cc79ba6c6c64db2e0a3b992da9f8). Getting myself a bit confused, but I probably should take a look at hail output from the current parallel.