Reverted filtering method for psms passed to FlashLFQ

[Alexander-Sol]

Psms were being filtered differently in ProteinScoringAndFdrEngine vs the QuantificationAnalysis method in PostSearchAnalysisTask, causing issues in protein quantification. This was caused by a switch to PEP based filtering in QuantificationAnalysis. This change was reverted, and should fix the issue.

[codecov[bot]]

Codecov Report

Merging #2261 (034d3cc) into master (6fcdc66) will decrease coverage by 0.01%. The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           master    #2261      +/-   ##
==========================================
- Coverage   91.44%   91.43%   -0.01%     
==========================================
  Files         134      134              
  Lines       20167    20158       -9     
  Branches     2809     2807       -2     
==========================================
- Hits        18441    18432       -9     
  Misses       1237     1237              
  Partials      489      489              

Impacted Files	Coverage Δ
...eus/TaskLayer/SearchTask/PostSearchAnalysisTask.cs	92.50% <100.00%> (-0.06%)	:arrow_down:

[Alexander-Sol]

Is PEP filtering implemented elsewhere?

Currently, filtering happens in ~12 different places. Sometimes by PEP, sometimes by Q-Value. In the future, this should be refactored so that a common filtering method can be employed. For right now though, I'm only changing enough to stop protein quant from breaking