Add support for cooperative groups and grid synchronization #6405

Job	Run time
clang-format	20s
linux_clang-14_cuda-11.2_debug_analysis	4m 40s
windows_cl-2022_debug_analysis	3h 25m 24s
macos_xcode-15.4_debug_analysis	17m 53s
linux_gcc-12_debug_analysis	22m 37s
macos_xcode-15.4_release	16m 36s
macos_xcode-16.1_debug	11m 10s
windows_cl-2022_release	2h 34m 10s
windows_cl-2022_debug	3h 20m 38s
linux_gcc-11_debug	21m 32s
linux_gcc-12_release_c++20	21m 32s
linux_gcc-13_debug	24m 19s
linux_clang-13_release	20m 3s
linux_clang-14_release	21m 29s
linux_clang-15_debug	15m 43s
linux_clang-16_debug_ubsan	19m 36s
linux_clang-16_debug_tsan	16m 47s
	13h 14m 29s

Provide feedback