New PySM 3 Galactic foregrounds simulation plan

[zonca]

Writing a plan for the next round of Simons Observatory simulations, with new foreground models and realistic bandpasses.

This is the first production run using the new PySM 3 native sky models, so it raises some general questions that affect all these kind of simulations and are not specific to SO.

TODO:

• Sigurd requested to switch CAR to fejer1 geometry in the fullsky_geometry function
• Implement output CAR in pysm and different modeling/output Nside Output car and modify Nside after smoothing galsci/pysm#125
• Prepare all input configuration
• Make sure HEALPix outputs are NESTED
• Make filenames compliant with SSIA doc
• if bandpass integration is slow, remove low points and decimate
• Run a couple of channels for testing
• Run production

[zonca]

Rendered PLAN.md (always points to the latest version): https://github.com/simonsobs/map_based_simulations/blob/202208_galactic_foregrounds/202208_galactic_foregrounds/PLAN.md

[zonca]

Looking for feedback on the plan, please do a "review" so you can add comments line by line into the document.

[zonca]

the only current concern about the models is about color-correcting the synchrotron templates: galsci/pysm#99

[zonca]

completed the production run on Popeye, working on verifying all the maps. Then need to create the combined output.