Set mutation positions using mutation time if available

[hyanwong]

Description

The first step to putting a y-axis on SVG plots.

Fixes #840

PR Checklist:

• Tests that fully cover new/changed functionality.
• Documentation including tutorial content if appropriate.
• Changelogs, if there are API changes.

[hyanwong]

This took a bit of reworking to include mutations above the root. But worth sorting as this will impact implementing a Y-axis scale, which we can now do by looking at node_y_coord_map and mut_y_coord_map. We should first implement a Y-scale on trees, and then generalise to tree sequence. This PR also includes a few nice example plots in tests/data/svg, e.g. the following for jumbled sample nodes

and this for a log time scale:

[jeromekelleher]

Looks good, although haven't delved into all the details. Getting full coverage in tests will be a faff.

[codecov]

Codecov Report

Merging #1230 (3becc79) into main (8f5bf4f) will increase coverage by 0.01%.
The diff coverage is 97.82%.

@@            Coverage Diff             @@
##             main    #1230      +/-   ##
==========================================
+ Coverage   93.72%   93.74%   +0.01%     
==========================================
  Files          26       26              
  Lines       21532    21583      +51     
  Branches      909      921      +12     
==========================================
+ Hits        20181    20232      +51     
- Misses       1312     1313       +1     
+ Partials       39       38       -1     

Flag	Coverage Δ
c-tests	92.50% <ø> (ø)
lwt-tests	92.97% <ø> (ø)
python-c-tests	94.93% <97.82%> (+0.02%)	⬆️
python-tests	98.63% <97.82%> (+0.01%)	⬆️

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

Impacted Files	Coverage Δ
python/tskit/drawing.py	99.04% <97.82%> (+0.07%)	⬆️

---

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 8f5bf4f...3becc79. Read the comment docs.

[hyanwong]

Looks good, although haven't delved into all the details. Getting full coverage in tests will be a faff.

What extra coverage would you like? E.g. should we check that a mutation high above a root is still within the tree box, and that the root branch is long enough? You can see this in the example SVGs, but it's not explicitly checked, which would be sensible, I guess. I'm not sure what else would be useful. Perhaps an example from the WF simulator for some multi-root trees, especially if we can make the JukeCantor routine place mutations at defined times?

[jeromekelleher]

My bad, looks like you have 100% coverage with the examples.

[hyanwong]

My bad, looks like you have 100% coverage with the examples.

OK, review comments done. Probably worth saving more extensive testing until we have y-scales, when it should be a bit more obvious where the mutations are, whether it's a log scale, etc. So maybe merge this if it's OK, and I'll get on with that.

[hyanwong]

Rolled into #1236 too, so we could either close this PR and review all the changes together over there, or merge this first and rebase. Either is fine by me.

[jeromekelleher]

Looks like a fair bit has changed in #1236, so will probably be reviewing twice if we go through it here too?

[hyanwong]

Looks like a fair bit has changed in #1236, so will probably be reviewing twice if we go through it here too?

Yeah, closing.