Refactor bbox to allow performance testing of transpose storage

[sethrj]

Many/most of the runtime bounding box algorithms loop over axes on the outside and bounds on the inside. This updates the bounding box internals to allow us to switch the axis/bound storage in the bounding box by writing most code as box.point(bound, axis).

• Add Extents3 type to allow construction/use by lo/hi pairs for each axis
• Use point(b, ax) in algorithms
• Add separate const ref and rvalue ref point accessors to allow ldg(&bbox.point(b, ax)) in future testing

[github-actions]

Test summary

 6 050 files   9 807 suites   19m 23s ⏱️
 2 311 tests  2 268 ✅  43 💤 0 ❌
34 432 runs  34 245 ✅ 187 💤 0 ❌

Results for commit c5d3a0d.

♻️ This comment has been updated with latest results.

[codecov]

Codecov Report

❌ Patch coverage is 99.09091% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 87.23%. Comparing base (097a308) to head (94bd2ae).

Files with missing lines	Patch %	Lines
src/geocel/BoundingBox.hh	98.18%	0 Missing and 1 partial ⚠️

Additional details and impacted files

@@             Coverage Diff             @@
##           develop    #2400      +/-   ##
===========================================
+ Coverage    87.22%   87.23%   +0.01%     
===========================================
  Files         1396     1396              
  Lines        44025    44050      +25     
  Branches     13791    13303     -488     
===========================================
+ Hits         38400    38427      +27     
+ Misses        4546     4397     -149     
- Partials      1079     1226     +147     

Files with missing lines	Coverage Δ
src/orange/BoundingBoxUtils.cc	100.00% <ø> (+2.27%)	⬆️
src/orange/BoundingBoxUtils.hh	97.10% <100.00%> (+0.20%)	⬆️
src/geocel/BoundingBox.hh	98.78% <98.18%> (+3.86%)	⬆️

... and 114 files with indirect coverage changes

🚀 New features to boost your workflow:

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

[elliottbiondo]

@sethrj did you look at timing for this?

[sethrj]

@elliottbiondo This should compile down to the same runtime code I think (some of the utils are slightly different but are only used in setup) so this one should have no performance impact. I'm waiting on ExCL to come back up to run the timing, and additionally check the timing of the bbox transpose.