Calibration of Q_pixel_size for rectangular pixels

[kofcla]

Hi everyone, thank you for this nice package!

Is your feature request related to a problem? Please describe.
I have been running into issues calibrating data that has rectangular pixels (e.g. datasets simulated with abTEM). Sometimes the sampling is different in x and y direction for reasons related to the simulations, so it would be great to have the possibility to give two values for the Q_pixel_size calibration.

Describe the solution you'd like
The possibility to calibrate the dataset with different values for the x and y Q_pixel_size.