Add E-scheme preprocessor fixing cluster history indices

[m-fila]

The ptscheme and pt2scheme preprocessors already assign correct history indices to jet objects. The default escheme does not currently have its own preprocessor: it either copies the full vector of jets when they are already PseudoJet/EEJet (assuming their cluster indices are correct), or it converts to PseudoJet/EEJet and assigns new history indices during that conversion.

As described in #200, relying on the input jets for escheme to already have correct cluster indices can be easy to overlook. Instead of performing checks and index fix-ups as suggested there, this proposal introduces a dedicated preprocessor for escheme that always copies jets and assigns correct history indices. This would become the default behavior for e-scheme (other scheme preprocessors already include fixing indices). Users who know their input jets already have valid history indices can opt out by specifying nothing as the preprocessor to avoid unnecessary work.

I also noticed that two forms of preprocessor functions exist: one that accepts an output type and performs conversion, and one that does not. The docs describes user-provided preprocessors as the non-converting type, but the implementation only uses the converting form. I fix the docs and remove the unused methods.

Finally, this PR includes fixes for history-index in SoftKiller (#191), as it is related and there has been no recent activity on that issue.

Closes #200 #191

[codecov]

Codecov Report

❌ Patch coverage is 85.71429% with 2 lines in your changes missing coverage. Please review.
✅ Project coverage is 80.01%. Comparing base (ae0ac95) to head (abd1d67).

Files with missing lines	Patch %	Lines
src/EEJet.jl	0.00%	2 Missing ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #211      +/-   ##
==========================================
- Coverage   80.22%   80.01%   -0.22%     
==========================================
  Files          21       21              
  Lines        1315     1326      +11     
==========================================
+ Hits         1055     1061       +6     
- Misses        260      265       +5     

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

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[m-fila]

Performance wise it seems calling the preprocess for fixing indices isn't more expensive than nothing preprocess just coping input jets without fixups:

• preprocess_escheme

  pp 14TeV Tiled
  Processed 100 events 16 times
  Average time per event 177.05440187500005 ± 18.650765441555805 μs
  Lowest time per event 161.72569 μs
  pp 14 TeV Plain
  Processed 100 events 16 times
  Average time per event 504.513385625 ± 18.07148126798454 μs
  Lowest time per event 486.39413 μs
  ee H Durham
  Processed 100 events 16 times
  Average time per event 32.36774750000001 ± 3.0916310772200886 μs
  Lowest time per event 29.67452 μs
  

• nothing

  pp 14TeV Tiled
  Processed 100 events 16 times
  Average time per event 176.36583312499997 ± 17.66222657424268 μs
  Lowest time per event 162.19935999999998 μs
  pp 14 TeV Plain
  Processed 100 events 16 times
  Average time per event 501.3887362499999 ± 16.77077525920156 μs
  Lowest time per event 485.47028 μs
  ee H Durham
  Processed 100 events 16 times
  Average time per event 35.764484375 ± 4.081166524662498 μs
  Lowest time per event 29.568640000000002 μs