Merge pull request #65 from mkstoyanov/better_fftw_threads_selection #157

Job	Run time
heffte (cmake, BUILTIN, cpu_intel)	1m 13s
heffte (cmake, FFTW, cpu_intel)	2m 7s
heffte (cmake, MKL, cpu_intel)	1m 50s
heffte (spack, BUILTIN, cpu_intel)	12m 6s
heffte (spack, FFTW, cpu_intel)	22m 31s
heffte (spack, MKL, cpu_intel)	12m 14s
heffte (cmake, ONEAPI, gpu_intel)	1m 32s
heffte (cmake, CUDA, gpu_nvidia)	7m 40s
heffte (spack, CUDA, gpu_nvidia)	18m 46s
heffte (cmake, ROCM, gpu_amd)	40s
	1h 20m 39s

Provide feedback