Reduce the size of the cuml libraries (#3702)

By explicitly telling nvcc's fatbin pass to always compress device code we can ensure that our binaries are the smallest possible size.

See rapidsai/cudf#7583 for additional context.

Authors:
  - Robert Maynard (https://github.com/robertmaynard)

Approvers:
  - Dante Gama Dessavre (https://github.com/dantegd)

URL: #3702

cpp/CMakeLists.txt

@@ -325,6 +325,9 @@ set(FAISS_GPU_ARCHS "${FAISS_GPU_ARCHS} ${ptx}")
 set(CMAKE_CUDA_FLAGS
   "${CMAKE_CUDA_FLAGS} -Xcudafe --diag_suppress=unrecognized_gcc_pragma")
 
+# make sure we produce smallest binary size
+string(APPEND CMAKE_CUDA_FLAGS " -Xfatbin=-compress-all")
+
 ##############################################################################
 # - dependencies -------------------------------------------------------------