PCT_URBAN=0 at 0° longitude using 0°-360° longitude mesh but works using -180-180 longitude mesh

### Brief summary of bug

I used mksurfdata_esmf for Europe (crossing 0° longitude) and found PCT_URBAN was incorrect using a 0-360 longitude mesh, but the other parameters seemed correct.
I also used a -180°-180° longitude mesh to generate the regional surface data, where the PCT_URBAN>0 at 0° longitude.

### General bug information

**CTSM version you are using:** ctsm5.3.021

**Does this bug cause significantly incorrect results in the model's science?** Not.
- Users could use 180°-180° longitude mesh as an alternative to generate correct surface data.
- PCT_URBAN=0 did not abort simulations.

**Configurations affected:** Not applicable

### Details of bug
PCT_URBAN=0 at 0° longitude when using 0°-360° longitude mesh.

### Important details of your setup/configuration so we can reproduce the bug

I compiled the mksurfdata_esmf using the GNU compiler on ARCHER2.
ESMF version:8.6.1
PIO version: 2.6.2

### Important output or errors that show the problem
Please see input and output files in https://www.dropbox.com/scl/fo/zrr05povz7fj21u21dlz8/AFMFt5pDCzyjopS2AFyTwoU?rlkey=wbun43dqua3q5q5uj99aru4ko&st=mgg4b2pc&dl=0