Import state nuopc state_diagnose call is not under dbug_flag > 1, causing ESMF logfile flood

### What happened?

In a run that has a preceding unrelated ESMF error (which creates the PET.. ESMF_LogFile), the run will get flooded with state diagnosis checks that take up CPU time and significant disk space

This is because this call
https://github.com/ESCOMP/CAM/blob/06cf015bb555fdcd2d2fc34e53bebe977230db86/src/cpl/nuopc/atm_comp_nuopc.F90#L1094-L1098

does not have a `dbug_flag > 1` gate like the others, e.g.,

https://github.com/ESCOMP/CAM/blob/06cf015bb555fdcd2d2fc34e53bebe977230db86/src/cpl/nuopc/atm_comp_nuopc.F90#L1185-L1190

one should be added to match the others

### What are the steps to reproduce the bug?

See above

the logfile will be flooded with stuff like this on every core
```
20260831 160442.916 INFO PET00 (med_methods_Field_GeomPrint):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_
anidr: dataptr bounds dim=1 1 144
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: distGrid=element
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: dimCount= 1
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: tileCount= 1
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: deCount= 96
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: localDeCount= 1
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: minIndexPTile= 1
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: maxIndexPTile= 13824
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: distGrid=nodal
20260831 160442.916 INFO PET00 (med_methods_Mesh_Print):(/home/hplin/260831_sima_cam4_fhist/CAM-SIMA.dev/components/cmeps/cime_config/../mediator/med.F90:completeFieldInitialization):Sl_avsdf
: dimCount= 1
```

### What CAM tag were you using?

cam6_4_199

### What machine were you running CAM on?

CISL machine (e.g. derecho)

### What compiler were you using?

Intel

### Path to a case directory, if applicable

_No response_

### Will you be addressing this bug yourself?

Any CAM SE can do this

### Extra info

assisted-by: claude-fable:5 when working on CAM-SIMA FHIST_C4
the equivalent bug is present in CAM-SIMA as well and will be fixed in another PR