dev: data reformatting plan (draft)

[nclack]

Adds a draft data-reformatting plan under dev/ (design/plan docs, outside the
mkdocs site), the actionable follow-on to the optimization report (docs/optimization.md).

For the ~9 decode-bound training datasets, it gives a target chunk/shard shape per
archetype (static 2D / video / 3D z-range) and the predicted decode waste after
re-chunking, with before→after computed from the real array+patch shapes:

• allencell-nuclear 131x -> 1.9x, allencell-static 128x -> 2.0x (target [1,1,1,256,256])
• the video MIPs (dynacell/jacobo/nucmorph/emt) 22-33x -> 2-4x
• raw 3D variants: their 76-163x is a 2D-crop-on-3D-data artifact; read with a
  z-range crop they're ~5-10x, and a small z-extent chunk takes them to ~2-5x

Principle: chunks small (64 KiB-1 MiB, smaller than the crop), with extent > 1 only
in dimensions that compress well together (x/y always, z in 3D, time sometimes,
channels split). Includes sharding guidance and verify-before-committing caveats.

Draft — predictions are from geometry; confirm one reformatted dataset with a real
bench run before doing the corpus.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 57.71%. Comparing base (cca3b09) to head (db1675f).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #158      +/-   ##
==========================================
- Coverage   57.78%   57.71%   -0.08%     
==========================================
  Files          64       64              
  Lines       10046    10055       +9     
  Branches     1748     1750       +2     
==========================================
- Hits         5805     5803       -2     
- Misses       3494     3501       +7     
- Partials      747      751       +4     

Flag	Coverage Δ
unittests	57.71% <ø> (-0.08%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.
see 5 files with indirect coverage changes

🚀 New features to boost your workflow:

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

[nclack]

Moved to docs/ in nclack/damacy-slides-2026-06; keeping these out of this repo.