Q value threshold adjustment and bug fix

[trishorts]

This PR changes the threshold used for using the more stringent and correct q-value calculation (D+1)/T to a total of 1000 PSMs. Below that level we use D/T. This will help with top-down files that contain fewer PTMs. It is hoped that in the future, we can eliminate this threshold and use one threshold for all. Some PRs that might help would be paired target-decoy and decoy on the fly using 10X decoys.

For reference on the equation please see:

J Proteome Res. 2022 Oct 7;21(10):2412-2420. doi: 10.1021/acs.jproteome.2c00282. Epub 2022 Sep 27.
https://pubmed.ncbi.nlm.nih.gov/36166314/

Traditional Q-Value (D/T) for PSM Count < 1000

[codecov]

Codecov Report

Attention: Patch coverage is 78.94737% with 4 lines in your changes missing coverage. Please review.

Project coverage is 93.63%. Comparing base (8c8fe5f) to head (d96c2da).
Report is 2 commits behind head on master.

Files with missing lines	Patch %	Lines
...pheus/EngineLayer/FdrAnalysis/FdrAnalysisEngine.cs	78.94%	0 Missing and 4 partials ⚠️

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #2426      +/-   ##
==========================================
- Coverage   93.67%   93.63%   -0.05%     
==========================================
  Files         141      141              
  Lines       21925    21940      +15     
  Branches     3007     3010       +3     
==========================================
+ Hits        20539    20544       +5     
- Misses        934      941       +7     
- Partials      452      455       +3     

Files with missing lines	Coverage Δ
...pheus/EngineLayer/FdrAnalysis/FdrAnalysisEngine.cs	93.54% <78.94%> (-4.04%)	⬇️

... and 2 files with indirect coverage changes

[Alexander-Sol]

the QvalueThresholdOverride property and associated field are a little wonky, I suggest changing it to a property with public get; private set;, then changing it in the relevant tests using reflection. Also, add the non-parallelizable tags