update the GFS_typesdef.meta in FV3

[XiaSun-Atmos]

Description

Add the active attribute to the vars wherever needed in the GFS_typesdef.meta file.
All the vars that are allocated in the GFS_typesdef.F90 based on if conditions were added an extra active attribute, except for phy_fctd.
In GFS_typesdef.F90:

    if (Model%nctp > 0 .and. Model%cscnv) then
      allocate (Tbd%phy_fctd (IM,Model%nctp))
      Tbd%phy_fctd = clear_val
    endif

When
active = (number_of_cloud_types_CS > 0 .and. flag_for_Chikira_Sugiyama_deep_convection)
is added to phy_fctd in GFS_typesdef.meta, error message is 'Exception: Variable number_of_cloud_types_cs used in conditional for cloud_base_mass_flux not known to host model

Further inspection reveals that this is related to the upper case used in the active attribute. Somehow items = FORTRAN_CONDITIONAL_REGEX.findall(var.active.lower()) in the mkstatic.py only returns lower case strings. I have tried items = FORTRAN_CONDITIONAL_REGEX.findall(var.active), then print (items), they are still lower case string.

Currently, the active attribute needed for phy_fctd is not included yet.

Testing

All 57 regression tests passed on Hera using the intel compiler with the current updated GFS_typesdef.meta file

Dependencies

NOAA-EMC/fv3atm#156

[climbfuji]

@XiaSun-NOAA Thanks for this PR. You will need to create a similar draft PR from within the FV3 subdirectory (i.e. for the fv3atm repository with the actual changes). Assuming you call the branch in fv3atm same as here, gfsmeta, you will need to update the ufs-weather-model .gitmodules file so that the fv3atm url points to your fork and the branch is gfsmeta.

[climbfuji]

Thanks, this looks great. I'll take a look at the changes in GFS_typedefs.F90 as soon as I can and let you know. I should be able to help you solve the problem you mentioned in the description.

[climbfuji]

This PR is no longer required, because NOAA-EMC/ufsatm#156 was pulled into NOAA-EMC/ufsatm#153. The submodule pointer update will happen in #180.

[climbfuji]

This PR is no longer required, because NOAA-EMC/ufsatm#156 was pulled into NOAA-EMC/ufsatm#153. The submodule pointer update happened in #180.