Add support for cooperative groups and grid synchronization #6410

Job	Run time
clang-format	20s
linux_clang-14_cuda-11.2_debug_analysis	4m 50s
windows_cl-2022_debug_analysis	32m 14s
macos_xcode-15.4_debug_analysis	15m 57s
linux_gcc-12_debug_analysis	22m 3s
macos_xcode-15.4_release	20m 22s
macos_xcode-16.1_debug	13m 54s
windows_cl-2022_release	32m 17s
windows_cl-2022_debug	32m 17s
linux_gcc-11_debug	20m 52s
linux_gcc-12_release_c++20	21m 24s
linux_gcc-13_debug	23m 17s
linux_clang-13_release	19m 25s
linux_clang-14_release	22m 25s
linux_clang-15_debug	14m 34s
linux_clang-16_debug_ubsan	18m 25s
linux_clang-16_debug_tsan	16m 32s
	5h 31m 8s

Provide feedback