subset_data: Fix conversion of Longitude to string

[samsrabin]

Description of changes

Resolves an issue in subset_data where Longitude types were being implicitly converted to string to look like <ctsm.longitude.Longitude object at 0x7f34cb836a90>. Now, implicit string conversion is disallowed—the user must specify the type of longitude they want.

Also adds testing:

• Python unit tests of implicit and explicit string conversion of Longitude objects
• Python system tests of string values in names of subset_data output files
• Python system testing that create_newcase works with subset_data outputs
• New CIME SystemTest SUBSETDATAPOINT. No; moving to a different PR.

Specific notes

Contributors other than yourself, if any: None

CTSM Issues Fixed:

• Resolves subset_data still having trouble with Longitude #3279

Are answers expected to change (and if so in what way)? No

Any User Interface Changes (namelist or namelist defaults changes)? No

Does this create a need to change or add documentation? Did you do so? No

Testing performed, if any: As of 319ea5a (worked at aca0a5d):

• Python unit tests
• Python system tests
• A manual run like so, approximating the setup @katierocci used:

outdir=$PWD/outdir_issue_3279
user_mods_dir=$outdir/user_mods
casedir=~/cases_ctsm/issue_3279

tools/site_and_regional/subset_data point --lat 45.402252 --lon 267.201915 --site ccc3_crujra --create-surface --create-datm --datm-syr 2000 --datm-eyr 2000 --create-user-mods --outdir $outdir --user-mods-dir $user_mods_dir

cime/scripts/create_newcase --case $casedir --res CLM_USRDAT --compset I2000Clm60BgcCrujra --run-unsupported --user-mods-dir $user_mods_dir

cd $casedir
./case.setup
./shell_commands
./preview_namelists
qcmd -- ./case.build && ./case.submit

• Another manual run like that, but using -92.798085 for the longitude to test whether it works for longitude type 180.

outdir=$PWD/outdir_issue_3279_type180
user_mods_dir=$outdir/user_mods
casedir=~/cases_ctsm/issue_3279_type180

tools/site_and_regional/subset_data point --lat 45.402252 --lon -92.798085 --site ccc3_crujra --create-surface --create-datm --datm-syr 2000 --datm-eyr 2000 --create-user-mods --outdir $outdir --user-mods-dir $user_mods_dir

cime/scripts/create_newcase --case $casedir --res CLM_USRDAT --compset I2000Clm60BgcCrujra --run-unsupported --user-mods-dir $user_mods_dir

cd $casedir
./case.setup
./shell_commands
./preview_namelists
qcmd -- ./case.build && ./case.submit

Also:

• clm_pymods suite: /glade/derecho/scratch/samrabin/tests_0627-161354de/

• As of 71f9995: New SystemTest: SUBSETDATAPOINT_Ld5_D_Mmpi-serial.CLM_USRDAT.I2000Clm60BgcCrujra.derecho_intel.clm-default. Completes successfully! Moving to a different PR to keep this one focused.

[slevis-lmwg]

@samsrabin and I went over this PR in detail today.
@samsrabin already documented a concern that I expressed, by opening issue #3293.