Support depth + normal map generation

[gizatt]

view_layers["View Layer"].use_pass_z
view_layers["View Layer"].use_pass_normal

And then somehow extract them. They get rendered alongside the combined image in the render result.

[gizatt]

https://blender.stackexchange.com/questions/56967/how-to-get-depth-data-using-python-api

Programmatically adding a viewer node for each channel seems like the right idea to force Blender to save those images in a way I can get at them.

This setup seems to spit out the right files (though you have to give each file output a different folder as a base path so the images don't overwrite each other -- how do you specify output file name?)