Optimize ACC port of atm_advance_scalars_mono_work:5555

[abishekg7]

This PR introduces optimizations for the OpenACC port of atm_advance_scalars_mono_work:5555. The table below lists the timings for a real, global 30km experiment on A100 GPU with the nvhpc.

For nvhpc gpu runs, -gpu=math_uniform is introduced as a build flag to ensure optimizations are bit-identical, and the we report the numbers using NV_ACC_TIME=1. The GPU runs are on 1 Derecho GPU node, using 1 A100 via 1 MPI task.

The numbers reported for the CPU runs with gnu and intel compilers use the newly-added timers local to this region, and are averaged across three runs. The CPU runs use a single derecho CPU node each, fully subscribed to 128 MPI tasks.

Version	GPU kernel time (ms)		CPU timer - gnu		CPU timer - intel25
base	6.095		0.007606		0.00773
1 - Split up into parallel regions	5.035		-		-
2 - Use vector collapse(2)	5.485		-		-
3 - Swap loop vector and seq in first parallel region	4.128		0.008813		0.008406
4 - Remove vector collapse(2) in 2nd & 3rd regions	3.684		-		-

The performance drop from base version to Opt 3 in the CPU case with GNU is 15.87% per invocation of this loop. But looking at the timings provided by atm_advance_scalars_mono, we actually see a 0.15% faster timing from base to Opt 3.

TODO: Add @Pranay-Reddy-Kommera as primary commit author

This work was completed in part at the NCAR/NLR/NOAA Open Hackathon, part of the Open Hackathons program. The authors would like to acknowledge OpenACC-Standard.org for their support.