Bug fixes for 32-bit physics & correct the lake scheme in FV3_HRRR_c3 & FV3_HRRR_gf

[SamuelTrahanNOAA]

Description

Corrects a few problems found in RRFS parallel development:

1. Wrong lake model in FV3_HRRR_gf and _c3
2. Uninitialized arrays.
3. Wrong real kind used during quilt server communication.

The first two fixes were from @SamuelTrahanNOAA and the last one is from @DusanJovic-NOAA

Issue(s) addressed

• fixes write grid component recover_fields can use the wrong real kind #689
• fixes clm lake and rrfs sd fields are not initialized before write grid component initialization #690
• fixes FV3_HRRR_c3 and FV3_HRRR_gf use the wrong lake scheme #691

Testing

How were these changes tested?

This is from a larger effort to get the 32-bit physics working with quilting restart. Unfortunately, we're not 100% successful yet. However, these fixes are urgently needed by the RRFS parallels, so you'll get the rest later.

What compilers / HPCs was it tested with?

Hera intel & gnu

Are the changes covered by regression tests? (If not, why? Do new tests need to be added?)

Yes. All 28 of the 13km conus tests are replaced with 16 new ones. There are new FV3_HRRR_gf and _c3 C96 regression tests.

Have the ufs-weather-model regression test been run? On what platform?

Yes. Hera.

Will the code updates change regression test baseline? If yes, why? Please show the baseline directory below.

28 tests are removed, 20 tests added; see ufs-weather-model PR

Please commit the regression test log files in your ufs-weather-model branch

Will do.

Dependencies

None.

[SamuelTrahanNOAA]

@tanyasmirnova should review this

[jkbk2004]

@junwang-noaa Can you review this pr? GSL team has urgency for their real-time experiment.

[zach1221]

@BrianCurtis-NOAA @DeniseWorthen we need one more approving review on this PR.