CHG: Avoid list comprehension for discrete select

[KatharineShapcott]

Changes Summary

• Discrete data selection speedup

Reviewer Checklist

• Are testing routines present?
• Do objects in the global package namespace perform proper parsing of their input?
• Are all docstrings complete and accurate?
• Is the CHANGELOG.md up to date?

[codecov]

Codecov Report

Base: 67.99% // Head: 67.97% // Decreases project coverage by -0.02% ⚠️

Coverage data is based on head (1753edf) compared to base (fe1d424).
Patch coverage: 88.88% of modified lines in pull request are covered.

Additional details and impacted files

@@            Coverage Diff             @@
##              dev     #427      +/-   ##
==========================================
- Coverage   67.99%   67.97%   -0.02%     
==========================================
  Files          82       81       -1     
  Lines       10004     9833     -171     
  Branches     2074     2032      -42     
==========================================
- Hits         6802     6684     -118     
+ Misses       2663     2610      -53     
  Partials      539      539              

Impacted Files	Coverage Δ
syncopy/datatype/base_data.py	76.48% <88.88%> (-0.25%)	⬇️
syncopy/plotting/_helpers.py	85.91% <0.00%> (-5.64%)	⬇️
syncopy/specest/freqanalysis.py	65.45% <0.00%> (-3.64%)	⬇️
syncopy/specest/compRoutines.py	91.24% <0.00%> (-1.35%)	⬇️
syncopy/plotting/mp_plotting.py	44.66% <0.00%> (-1.34%)	⬇️
syncopy/statistics/compRoutines.py	81.42% <0.00%> (-0.89%)	⬇️
syncopy/statistics/summary_stats.py	75.44% <0.00%> (-0.61%)	⬇️
syncopy/statistics/jackknifing.py
syncopy/tests/__init__.py	65.45% <0.00%> (+0.31%)	⬆️
... and 1 more

Help us with your feedback. Take ten seconds to tell us how you rate us. Have a feature suggestion? Share it here.

☔ View full report at Codecov.
📢 Do you have feedback about the report comment? Let us know in this issue.

[tensionhead]

Thx @KatharineShapcott, looks like a nice and clean improvement! Probably the biggest speedup comes from harnessing the new _trialslice and channel_idx?! However employing list comprehensions to filter data directly definitely was not a good idea in any case..

[KatharineShapcott]

Yeah I think that was the biggest, but for looping through trialArr multiple times per trial was also not good when it can include 50,000 spikes per trial!