Weak lensing: decouple from master, refactor descwl/anacal workflow

[timwhite0]

• Separated WeakLensingNet (which worked for both DC2 and descwl) into MassMapEncoder (DC2) and ScalarShearEncoder (descwl).
• Created standalone versions of variational_dist, cached_dataset, dc2, utils to eliminate reliance on master.
• Simplified descwl and AnaCal workflow. Previously separate versions of the images were stored for NPE and AnaCal, now a single version of each simulation setting is stored in /nfs/turbo. Some steps for generating the final figures for each setting were not previously committed; now we demonstrate the full workflow in descwl/results.
• Reorganized and renamed configs and notebooks in the dc2 and descwl subfolders. Ran all notebooks. Old notebooks and scripts are marked as deprecated.
• Rewrote README with updated usage instructions.

Our code for generating descwl-shear-sims images is in the image_generation folder in our descwl-shear-sims fork.

Future development of this case study will happen in blissWL.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 80.64%. Comparing base (39faef5) to head (18d50bb).
⚠️ Report is 1 commits behind head on master.

Additional details and impacted files

@@           Coverage Diff           @@
##           master    #1104   +/-   ##
=======================================
  Coverage   80.64%   80.64%           
=======================================
  Files          22       22           
  Lines        3111     3111           
=======================================
  Hits         2509     2509           
  Misses        602      602           

Flag	Coverage Δ
unittests	80.64% <ø> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ 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.