Put some order in our backends, options, and architectures
Fix #370
Grouping backends by serial, host parallel, and device
Grouping architectures by CPU vs GPU and subdivide based on vendors
I focused on the NVIDIA GPUs. AMD and Intel GPUs can be tackled in follow ups.
CPUs could use some TLC too.
This can obviously be improved further but I think it is already worth merging.
Put some order in our backends, options, and architectures Fix #370
Grouping backends by serial, host parallel, and device
Grouping architectures by CPU vs GPU and subdivide based on vendors I focused on the NVIDIA GPUs. AMD and Intel GPUs can be tackled in follow ups. CPUs could use some TLC too.
This can obviously be improved further but I think it is already worth merging.