add num_children Python

[GertjanBisschop]

First pass at low level Python for tree.num_children_array and Python API. Was mostly copy and paste as expected, but still second-guessing myself I might have missed something.
I have set the type of the num_children_array to UINT64 (see tsk_size_t). Hope this is the correct 'translation'.

[codecov]

Codecov Report

Merging #2332 (70d7e1c) into main (7b66382) will increase coverage by 0.01%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##             main    #2332      +/-   ##
==========================================
+ Coverage   93.28%   93.29%   +0.01%     
==========================================
  Files          28       28              
  Lines       26722    26730       +8     
  Branches     1222     1222              
==========================================
+ Hits        24927    24939      +12     
+ Misses       1762     1758       -4     
  Partials       33       33              

Flag	Coverage Δ
c-tests	92.25% <100.00%> (ø)
lwt-tests	89.05% <ø> (ø)
python-c-tests	71.40% <92.30%> (+0.03%)	⬆️
python-tests	98.85% <100.00%> (+<0.01%)	⬆️

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

Impacted Files	Coverage Δ
c/tskit/trees.c	95.08% <100.00%> (ø)
python/_tskitmodule.c	90.74% <100.00%> (+0.06%)	⬆️
python/tskit/trees.py	98.28% <100.00%> (+<0.01%)	⬆️

---

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 7b66382...70d7e1c. Read the comment docs.

[jeromekelleher]

LGTM!

The UINT64 thing raises an important question, which I've opened #2333 to discuss.

[benjeffery]

Looks good, we might want to add num_children to data-model.md as that documents the other arrays?

Also might be nice to simplify things like test_topology:239

I think num_children needs to be added at test_topology:1793,1825,1880,1914,2185,6482

[benjeffery]

Also, a changelog would be appreciated!

[GertjanBisschop]

Had a go at adding in some documentation on the convenience arrays as I tentatively called them, based on @benjeffery's suggestion. Have kept them separate from the paragraph on the quintuply linked tree to split off the essential from the non-essential.

[benjeffery]

Once #2337 is merged we can modify this PR to return the 32bit type.

[benjeffery]

@jeromekelleher I've added a commit here to convert to int32. Would you do a quick sanity review of that before we merge here?

[jeromekelleher]

LGTM!