Data description update

[s0larish]

PR summary

Update the documentation to highlight which dataproducts to be used primarily for scientific analysis.
Completes #687

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 57.39%. Comparing base (185054f) to head (cebc34d).
⚠️ Report is 39 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #925      +/-   ##
==========================================
- Coverage   57.40%   57.39%   -0.01%     
==========================================
  Files         125      125              
  Lines       13885    13887       +2     
==========================================
  Hits         7970     7970              
- Misses       5915     5917       +2     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[lowderchris]

I might flip this throughout the documentation - so refer to them first as clear / polarized low-noise mosaics, with CAM / PAM in parentheses? Just to avoid getting lost in an alphabet soup of product codes for users.

[svank]

But on the other hand, if we focus on "polarized" and "clear", that might distract from the fact that PAM also includes unpolarized data.

[jmbhughes]

Suggested change

	Both are fully background and starfield subtracted and mosaicked across the PUNCH field of view,
	Both products are fully background and starfield subtracted and mosaicked across the PUNCH field of view,

[svank]

But on the other hand, if we focus on "polarized" and "clear", that might distract from the fact that PAM also includes unpolarized data.

[svank]

Suggested change

	For most science use cases, the recommended starting point is the Level 3 **CAM**
	(Clear a.k.a total brightness low-noise science mosaic) and **PAM** (Polarized low-noise science mosaic) products.
	Both are fully background and starfield subtracted and mosaicked across the PUNCH field of view,
	giving you the cleanest possible signal for heliospheric science.
	CAM provides total brightness (B) with uncertainty, while PAM additionally resolves
	polarized brightness (pB) and its radial component (pB'), making PAM the product of
	choice for studies of the CMEs, shocks, solar wind etc. These products
	represent one full spacecraft rotation cycle with a significant
	improvement in signal-to-noise relative to other products.
	For most science use cases, the recommended starting point is the two Level 3 low-noise science mosaic products,
	**PAM** and **CAM**.
	Both products are fully background and starfield subtracted and mosaicked across the PUNCH field of view,
	giving you the cleanest possible signal for heliospheric science.
	Each of these two product types are produced every 32 minutes, allowing multiple images to be averaged together at every point in the mosiac field of view,
	offering improved signal-to-noise relative to other products.
	PAM files (with the "P" standing for "polarized") have three data layers: total brightness (tB), which is effectively an unpolarized image, followed by
	polarized brightness (pB) and its radial component (pB'), making PAM the product of
	choice for studies of CMEs, shocks, the solar wind, etc.
	CAM files (with the "C" standing for "clear") provide only unpolarized data. The PAMs' unpolarized total brightness is reconstructed from sets of polarized observations, while the CAMs are produced from unpolarized observations which are made at a reduced cadence. So while the total brightness layer of the PAMs is conceptually equivalent to the CAMs, PAMs average together more data, which is expected to provide better signal-to-noise. But for memory- or bandwidth-constrained situations that don't require polarization, CAMs provide largely equivalent data at one-third the file size.

Here are some possible additions to this paragraph. I think two product codes aren't enough to be alphabet soup, and I think their differences and similarities are subtle enough to deserve a few more sentences. I think especially the fact that PAMs include "clear" data, and the differences between that and the CAMs, deserves careful attention and emphasis.