simplify filter_variance_skimage_napari.py

[tischi]

I was trying to reproduce our code in filter_variance_skimage_napari.py but failed.

Something was wrong with the computation of the variance image, it looked always very noisy and thus I could not threshold it successfully.

I changed this now.

Even though it introduces the concept of a function I find it much more readable, robust and useful than the previous approach, where a lot of code and mental load went into how to compute a variance; and, in fact, it did not work, for me.

I would have no worries that the students would understand this, because the function is super simple; and it actually very nicely fits the concept map of that you do some local math in a region of the image.

[tischi]

I need to teach this tomorrow, thus please comment today; sorry for the rush.

[manerotoni]

The function is very simple and the explanation is clear. I wonder if one should just plainly run it once on a chunk of image unit16_image[0:10,0:10].
This will help understanding the new concept of generic_filter