Remove teaching the LayerDataTuple in the napari exercise

[haesleinhuepf]

We should consider removing teaching the LayerDataTuple-exercise for three reasons:

• It's not necessary for a first napari plugin
• It's confusing
• Functions that return a LayerDataTuple are harder to reuse from Jupyter.

I would remove this notebook: https://github.com/BiAPoL/Quantitative_Bio_Image_Analysis_with_Python_2022/blob/main/docs/day4c_crafting_napari_plugins/02_creating_widgets_from_functions_part2.ipynb

Instead it would be good if we can teach how to make functions that are resuable as napari plugin and as Jupyter notebook. Example:

def my_workflow(image:"napari.types.ImageData", sigma:float=1) -> "napari.types.LabelsData":

    # ...

    return label_image

[zoccoler]

I removed contents referring to LayerDataTuple and removed the mentioned notebook. 72a6aef1aec5a5c43b973708c1b21f3f5017de35

I kept importing napari data types because they seem to be more reliable (check this comment)

[haesleinhuepf]

Thanks for fixing this and thanks for the pointer! I'm generally using the "napari.types.ImageData" strings now. Maybe, we create some teaching materials for a future course where we can show LayerDataTuple, napari.types.ImageData np.ndarray, and others. I think it might make sense to elaborate on the pros and cons of these different strategies. And I hope that there will be an advanced course on these things.

Thanks again!