fix: prefer NVML v2 memory info for inference setup#41

foreverlms merged 4 commits into

NVIDIA:mainNVIDIA/cosmos-framework:mainfrom

Rohithmatham12:fix-nvml-v2-memory-infoRohithmatham12/cosmos-framework:fix-nvml-v2-memory-infoCopy head branch name to clipboard