Upgrade to spack-stack/1.6.0

[DavidHuber-NOAA]

DESCRIPTION OF CHANGES:

This upgrades the libraries for UFS_Utils to those of spack-stack version 1.6.0. Most notably, this upgrades the following libraries:

• netcdf-fortran 4.6.0 -> 4.6.1
• esmf 8.4.2 -> 8.6.0
• sp 2.3.3 -> 2.5.0

Also, support for Cheyenne is removed as that system has gone offline.

Regression tests on Hera, Orion, Hercules, and Jet all showed the same differences

• Minor differences in U- and V-winds in several chgres tests. @GeorgeGayno-NOAA confirmed these were minor.
• The mx100 cpld_gridgen test showed differences in col and S variables. @DeniseWorthen confirmed these were minor and had to do with the way ESMF regridded the variables.

One test differed on Orion/Hercules, Jet, and WCOSS2:

• The global_cycle C192 landinc soil-NoahMP test produced identical surface .inc files on Hera, but on Orion/Hercules, Jet and
  WCOSS2 the tiled sfcanl files had an extra checksum attribute in the hice and fice fields. However, the values of the fields were identical. This probably means the baseline data needs to be updated on these platforms.

TESTS CONDUCTED:

If there are changes to the build or source code, the tests below must be conducted. Contact a repository manager if you need assistance.

• Compile branch on all Tier 1 machines using Intel (Orion, Jet, Hera, Hercules and WCOSS2).
• Compile branch on Hera using GNU.
• Compile branch in 'Debug' mode on WCOSS2.
• Run unit tests locally on any Tier 1 machine.
• Run relevant consistency tests locally on all Tier 1 machine.

Describe any additional tests performed.

ISSUE:

#884

[GeorgeGayno-NOAA]

There is a reference to cheyenne in machine_setup.sh that you can remove:
https://github.com/DavidHuber-NOAA/UFS_UTILS/blob/ss_160/sorc/machine-setup.sh#L40

[GeorgeGayno-NOAA]

The global_cycle regression test baseline was recently updated under a previous PR. So, the 'checksum' difference is no longer an issue. When I run your branch (4e6757a) on Hera, all global_cycle tests pass.

[DavidHuber-NOAA]

There is a reference to cheyenne in machine_setup.sh that you can remove: https://github.com/DavidHuber-NOAA/UFS_UTILS/blob/ss_160/sorc/machine-setup.sh#L40

Done.

The global_cycle regression test baseline was recently updated under a previous PR. So, the 'checksum' difference is no longer an issue. When I run your branch (4e6757a) on Hera, all global_cycle tests pass.

Noted, thanks for testing.

[GeorgeGayno-NOAA]

'develop' was updated this morning. Did you merge these changes to your branch?

[DavidHuber-NOAA]

No, but I have now.

[GeorgeGayno-NOAA]

Looks good. Will merge.