Update np.float to float
numpy no longer provides np.float, and the plain float type should be used instead. Newer numpy versions now raise an error for this, so I'm updating the code. This MR also includes version bump and changelog for a point release