PTI example jupyter notebook to script and .html

[talonchandler]

This PR moves the PTI spoke-pattern data example from a jupyter notebook to a script and a static .html file.

This PR also includes several suggestions from @mattersoflight to match the names to the most recent PTI submission:

• [x] Change section title 3D uPTI reconstruction -> 3D PTI reconstruction.
• [x] Change subsection title Reconstructed phase, absorption, principal retardance, azimuth, and inclination assuming (+) and (-) optic sign -> Reconstructed mean permittivity, differential permittivity, azimuth, and inclination assuming (+) and (-) optic sign. The text should clarify that the real part of mean permittivity causes phase delay and imaginary part of mean permittivity causes absorption.
• [x] Figure title for phase (xz) reads absorption (xz).
• [x] Figure title principal retardance -> differential permittivity
• [x] @ziw-liu I've applied black to the notebook-style script

Some other TODOs/questions:

• [x] Decide if we want a .pdf instead of .html. My first attempt at a .pdf led to TexLive issues together with font- and figure-size weirdness. The .html worked well out of the box. I'm open to suggestions and constraints I'm not thinking of.
• [x] Delete the jupyter notebook before merging. I've kept it temporarily since it's the source of the other two files and I may need to change them all together in this PR.

[talonchandler]

One disadvantage of the .html file is that github doesn't render it, and there doesn't seem to be an easy way to open it in another tab. For comparison here's a saved .pdf of the .html file.
PTI_Experiment_Recon3D_anisotropic_target_small.pdf

[ziw-liu]

GitHub doesn't render it because it's too large. But we can link to a third party one e.g. https://html-preview.github.io/?url=https://github.com/mehta-lab/waveorder/blob/a93fd512c6d66cb1b4bd8b983b288195cbb67fe5/examples/documentation/PTI_experiment/PTI_Experiment_Recon3D_anisotropic_target_small.html

[mattersoflight]

@talonchandler @ziw-liu thanks! Are you using vscode's Jupyter extension for PDF export? Does it require TexLive in the backend?

I get the following PDF after running two cells. test.pdf

I will be able to do the proper review of the script tomorrow and can export a more complete PDF in the process. @ziw-liu's solution of rendering via https://html-preview.github.io/ also works. We can link the HTML in the README via that service.

[talonchandler]

Aha, thanks. I got jupyter nbconvert (what vscode's convert uses) to run (I needed conda install pandoc) and the pdf looks good. I think I prefer the pdf over the html, so I updated to the pdf.

[talonchandler]

Thanks Shalin:

• [x] reconstructions should save to napari-readable zarr with channel names (see attached screenshot)
• [x] principal retardance should be differential permittivity (thanks, pardon my mistake)
• [x] don't track the jupyter notebook

[talonchandler]

P.S. I tested that all of the cells run...only thing that needs to change is the path to the data.

[talonchandler]

Thanks for the fixes! Merging and releasing now.