some issues when build with win10 use clion cuda11.1 tensorrt 8.0

[lqyp0420]

batchedNMSPlugin/kernel.h line 32 std::max->std::fmax; batchedNMSPlugin batchedNMSInference.cpp and batchedNMSPlugin.h nmsInference is not the same; torchCumPlugin torchCumPlugin.cpp line 58 or->||;

cmakeslist

1. find_package(CUDA REQUIRED) changed to set(CMAKE_SIZEOF_VOID_P 8) FIND_PACKAGE(CUDA REQUIRED);
2. add libraries when use cublas:target_link_libraries(${SHARED_TARGET or STATIC_TARGET} ${CUDA_cublas_LIBRARY})

[grimoire]

Thanks for the notification! Would you like to create a PR to fix it?