[libc] Fix using the libcgpu.a for NVPTX in non-LTO builds

CUDA requires a PTX feature to be compiled generally, because the
`libcgpu.a` archive contains LLVM-IR we need to have one present to
compile it. Currently, the wrapper fatbinary format we use to
incorporate these into single-source offloading languages has a special
option to provide this. Since this was not present in the builds, if the
user did not specify it via `-foffload-lto` it would not compile from
CUDA or OpenMP due to the missing PTX features. Fix this by passing it
to the packager invocation.

Reviewed By: jdoerfert

Differential Revision: https://reviews.llvm.org/D154864

Author: jhuber6

libc/cmake/modules/LLVMLibCObjectRules.cmake

@@ -184,8 +184,14 @@ function(_build_gpu_objects fq_target_name internal_target_name)
 
       # Append this target to a list of images to package into a single binary.
       set(input_file $<TARGET_OBJECTS:${gpu_target_name}>)
-      list(APPEND packager_images
-           --image=file=${input_file},arch=${gpu_arch},triple=${gpu_target_triple})
+      if("${gpu_arch}" IN_LIST all_nvptx_architectures)
+        string(REGEX MATCH "\\+ptx[0-9]+" nvptx_ptx_feature ${nvptx_options})
+        list(APPEND packager_images
+             --image=file=${input_file},arch=${gpu_arch},triple=${gpu_target_triple},feature=${nvptx_ptx_feature})
+      else()
+        list(APPEND packager_images
+             --image=file=${input_file},arch=${gpu_arch},triple=${gpu_target_triple})
+       endif()
       list(APPEND gpu_target_names ${gpu_target_name})
     endforeach()