Add support for cooperative groups and grid synchronization #6406

Job	Run time
clang-format	17s
linux_clang-14_cuda-11.2_debug_analysis	4m 26s
windows_cl-2022_debug_analysis	2h 56m 58s
macos_xcode-15.4_debug_analysis	24m 4s
linux_gcc-12_debug_analysis	21m 48s
macos_xcode-15.4_release	25m 32s
macos_xcode-16.1_debug	20m 10s
windows_cl-2022_release	2h 34m 13s
windows_cl-2022_debug	2h 56m 58s
linux_gcc-11_debug	20m 23s
linux_gcc-12_release_c++20	21m 39s
linux_gcc-13_debug	27m 24s
linux_clang-13_release	20m 1s
linux_clang-14_release	22m 42s
linux_clang-15_debug	14m 30s
linux_clang-16_debug_ubsan	18m 17s
linux_clang-16_debug_tsan	16m 35s
	12h 45m 57s

Provide feedback