Tighten pixelLessStep filter

[mmasciov]

PR description:

As per title, after PR #381: tightening of pixelLessStep filter in endcaps & at low pT.

PR validation:

• TTbar with PU: https://mmasciovecchio.web.cern.ch/MkFit_hitSelectionWindows_Nov21/MTV_TTbar_PU_PR382_12X/plots_pixelLessStep.html

• High-pT QCD without PU: https://mmasciovecchio.web.cern.ch/MkFit_hitSelectionWindows_Nov21/MTV_highPtQCD_noPU_PR382_12X/plots_pixelLessStep.html

--> Mild loss of efficiency in endcaps & at low pT, with significant reduction of fakes.

As a result, OOB high-purity tracks are back under control:

• TTbar with PU: https://mmasciovecchio.web.cern.ch/MkFit_hitSelectionWindows_Nov21/MTV_TTbar_PU_PR382_12X/plots_highPurity/effandfakePtEtaPhi.pdf

• High-pT QDC without PU: https://mmasciovecchio.web.cern.ch/MkFit_hitSelectionWindows_Nov21/MTV_highPtQCD_noPU_PR382_12X/plots_highPurity/effandfakePtEtaPhi.pdf