build

ACC: use data regions rather than hard-coded parallel gang loops #643

Job	Run time
ci (Release, DP, linux gnu-10)	32m 35s
ci (Release, DP, linux gnu-14)	32m 13s
ci (Release, DP, linux nvhpc-21.9)	32m 33s
ci (Release, DP, linux intel-classic)	16m 32s
ci (Release, DP, macos)	34m 28s
ci (Release, SP, linux gnu-10)	32m 24s
ci (Release, SP, linux gnu-14)	31m 9s
ci (Release, SP, linux nvhpc-21.9)	33m 27s
ci (Release, SP, linux intel-classic)	11m 36s
ci (Release, SP, macos)	27m 11s
	4h 44m 8s

Provide feedback