[OpenMP] NVPTX: Set default/minimum compute capability to sm_35

The current implementation of the nvptx runtime (to be upstreamed shortly) uses the atomicMax operation on 64-bit integers.
This is only supported in compute capabilities 3.5 and later. I've changed the clang default to sm_35.

Differential Revision: https://reviews.llvm.org/D40977



git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@320082 91177308-0d34-0410-b5e6-96231b3b80d8