Making PGFPlotsX use deterministic series IDs

[ajdunlap]

Description

This is a first attempt at addressing #4927 . One downside I can see is that this may impact thread-safety. It also may be a bit slower if there are many many series, but it seems like there would have to be really a lot of series for the maximum call to take a significant amount of time.

Attribution

• I am listed in .zenodo.json (see How to get acknowledged #3503)

Things to consider

• Does it work on log scales?
• Does it work in layouts?
• Does it work in recipes?
• Does it work with multiple series in one call?
• PR includes or updates tests?
• PR includes or updates documentation?

[codecov]

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 89.69%. Comparing base (fa65e7d) to head (2b89cbe).
Report is 13 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4932      +/-   ##
==========================================
- Coverage   89.77%   89.69%   -0.08%     
==========================================
  Files          40       40              
  Lines        8780     8792      +12     
==========================================
+ Hits         7882     7886       +4     
- Misses        898      906       +8     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[BeastyBlacksmith]

The problem I see here is, that when you have multiple such outputs in a larger tex-file you really want the ids to be unique to the whole document and not just to a single plot, or a particular process.

[ajdunlap]

I just tried a test file with two identical PGFPlots with identical "name path" values and it compiled correctly and didn't complain about the repeated "name path" values. Of course it also doesn't complain when there are repeated name path values in the same plot, but then of course the output is not correct. But if it is essential to have unique name paths even through multiple plots in the same document, then indeed something more complicated is required here.

[BeastyBlacksmith]

I'm okay with this then, I think this should be the only use of UUIDs in the codebase and thus we can remove this dependency then.
Can you also open a PR with this against the v2 branch?