In our FFTW wrapper in WarpX/ABLASTR, we support threads.
It seems AMReX does not support OpenMP/threads with FFTW yet.

ABLASTR: https://github.com/BLAST-WarpX/warpx/blob/25.12/Source/ablastr/math/fft/WrapFFTW.cpp#L44-L52
AMReX: https://github.com/AMReX-Codes/amrex/blob/25.12/Src/FFT/AMReX_FFT_Helper.H#L285-L304

This causes simulations with few/single boxes to not make use of CPU threads:
Re: https://github.com/orgs/BLAST-ImpactX/discussions/1245