Add NVML support for persistence mode, locking clocks.

Locking clocks is currently only implemented for Volta+ devices.

Example usage:

my_bench -d [0,1,3] --persistence-mode 1 --lock-gpu-clocks base

See the cli_help.md docs for more info.

cmake/NVBenchDependencies.cmake

@@ -57,3 +57,10 @@ rapids_find_package(CUDAToolkit REQUIRED
 
 # Append CTK targets to this as we add optional deps (NMVL, CUPTI, ...)
 set(ctk_libraries CUDA::toolkit)
+
+################################################################################
+# CUDAToolkit -> NVML
+if (NVBench_ENABLE_NVML)
+  include("${CMAKE_CURRENT_LIST_DIR}/NVBenchNVML.cmake")
+  list(APPEND ctk_libraries nvbench::nvml)
+endif()