Refactor law of motion.

[tobiasraabe]

• Refactor the law of motion such that the whole logic is only in one place.
• Simplify the collection of child indices.

[codecov]

Codecov Report

Merging #373 into main will increase coverage by 0.46%.
The diff coverage is 86.27%.

@@            Coverage Diff             @@
##             main     #373      +/-   ##
==========================================
+ Coverage   85.42%   85.89%   +0.46%     
==========================================
  Files          47       47              
  Lines        3411     3410       -1     
==========================================
+ Hits         2914     2929      +15     
+ Misses        497      481      -16     

Flag	Coverage Δ
#end_to_end	75.24% <84.31%> (+0.46%)	⬆️
#integration	65.42% <84.31%> (+0.54%)	⬆️
#unit	70.05% <86.27%> (+0.57%)	⬆️

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

Impacted Files	Coverage Δ
respy/shared.py	82.81% <73.68%> (+0.95%)	⬆️
respy/_numba.py	87.17% <80.00%> (-1.06%)	⬇️
respy/state_space.py	93.86% <94.73%> (+5.81%)	⬆️
respy/simulate.py	93.56% <100.00%> (-0.34%)	⬇️
respy/tests/test_simulate.py	100.00% <100.00%> (ø)
respy/interpolate.py	96.34% <0.00%> (+1.21%)	⬆️

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update ddde5de...c5b44f7. Read the comment docs.

[mo2561057]

Great PR. Very useful to recognize that the LOM and the child index creation use the same logic!

[mo2561057]

Would it be an option to not apply the law of motion here but take choice and dense key to get the next indices directly from the state space?
Then we would also save the indexer step and save the calculation.

[tobiasraabe]

I have created the issue #376 to discuss your point. Even if my point on cpu bound and io bound is not true, it makes sense to separate the changes. Is it ok with you?