Feature description

Add support for a 64 bit buffer (i.e. double).

Additional context

https://gdal.org/en/stable/doxygen/gdal__alg_8h.html#a7a789015334d677afcbef67e5a6b4a7c

The source pixel band values are read into a 32bit float buffer.

I recently ran into a scenario where a numpy.float64 value was truncated and encountered this in the docs.