feat: Python particle gun and Fatras examples can be chained

[timadye]

First step implementing #1098, incorporating #1093.

• addParticleGun and addFatras adds the required modules to the Sequence.
• runParticleGun and runFatras rewritten to use add* to reproduce exactly the previous behaviour.
• ROOT_HASH_CHECKS=on pytest run as before without change.
  • Note that ROOT_HASH_CHECKS=on with pytest doesn't currently work from main. I have a couple of fixes for future PRs.
• Optionally, runParticleGun and runFatras (like add*) can now accept outputDir as pathlib.Path as well as str. This should be the only change to how these routines work.

What do people think? If this is the right way to go, I can do the other Python examples - either in this PR or as a separate one if this one is merged.

[codecov]

Codecov Report

Merging #1128 (6e255a1) into main (853f9b7) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##             main    #1128   +/-   ##
=======================================
  Coverage   47.90%   47.90%           
=======================================
  Files         359      359           
  Lines       18491    18491           
  Branches     8723     8723           
=======================================
  Hits         8859     8859           
  Misses       3603     3603           
  Partials     6029     6029           

---

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 853f9b7...6e255a1. Read the comment docs.

[timadye]

I have removed the WiP label. Maybe the WiP label has inhibited people from looking. This proposal has been complete (as far as it goes) and ready for review. I have not been working on it, but used the label originally because we could expand on it later.

[paulgessinger]

I think I like this. I added some discussion points, but overall this feels like the right direction to me.

Should we add some form of example how to do the chaining in this PR already? Maybe in the docs?

[timadye]

Should we add some form of example how to do the chaining in this PR already? Maybe in the docs?

There is an example in runFatras. I'd prefer to hold off making docs until the other code is modified, so we can refer to more examples.

[paulgessinger]

Fair enough @timadye.

[timadye]

arrrgggg, GitHub foiled me again with its crazy-ass UI. I had been making all these comments and replies (one even before the holidays), but they all remained "Pending". I didn't spot that until just now and couldn't see what to do about it until I found that I had to go to the "Changes" section and click the "Review" button. Why oh why oh why....

[paulgessinger]

About namedtuple and defaults: yeah it's more verbose, that's true. I think it should be possible to construct a namedtuple without specifying the field names, and I think there's also a way to supply defaults. Another alternative in the same vein would be dataclasses, I guess.

[robertlangenberg]

Thanks Tim, the changes look good!