FINUFFT_EXECUTE architecture dispatch

[DiamonDinoia]

It is possible to compile the code in finufft_execute for multiple SIMD instructions and select the fastest available at runtime. This might not impact power users (that compile the code themselves) but it will impact python/matlab users are well as developers that cannot use -march=native.

Once this is merged we might consider shipping an linux artifacts on package managers (APT and similars) as these will run with good perfomance everywhere.