Fix errors in CMake CUDA flags and set the NVCC language level to 14

[alexrlongne]

Background

• I didn't fully test some of the CMake for the new CUDA specific device code and I had a few syntax errors in CMake. This pull request corrects those errors and adds a language level to the nvcccompiler, set through theCMAKE_CUDA_STANDARD` variable.

Purpose of Pull Request

• Fixes bug in CUDA specific CMake code

Description of changes

• Fix set statements that were not terminated properly
• Add FORCE to the set statement for CMAKE_CUDA_FLAGS
• Set the CMAKE_CUDA_STANDARD to 14

Status

• Reference:
  • Pre-Merge Code Review
  • CDash Status
• Checks
  • Travis CI checks pass
  • Code coverage does not decrease
  • Valgrind test passes
  • Toss3 checks pass
  • Trinitite checks pass
  • Code reviewed/approved, sufficient DbC checks, testing, documentation

[codecov]

Codecov Report

Merging #604 into develop will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           develop    #604   +/-   ##
=======================================
  Coverage     93.7%   93.7%           
=======================================
  Files          381     381           
  Lines        17805   17805           
=======================================
  Hits         16687   16687           
  Misses        1118    1118

[codecov]

Codecov Report

Merging #604 into develop will increase coverage by 1.6%.
The diff coverage is n/a.

@@            Coverage Diff            @@
##           develop    #604     +/-   ##
=========================================
+ Coverage       92%   93.7%   +1.6%     
=========================================
  Files          368     381     +13     
  Lines        17073   17805    +732     
=========================================
+ Hits         15714   16687    +973     
+ Misses        1359    1118    -241