Toggle to disable the pLS duplicate cleaning

This PR introduces a toggle (-2) to disable pLS duplicate cleaning. This is not meant to be a proposed change, as it both increases the timing and has dramatic physics performance changes, but it serves as a technical change to facilitate more configuration when including Patatrack pixel tracks as input in the future.

This is the master version of the now closed PR #318, and is only possible for merging after the updates in PR #324.

Timing

Before (48e470e0b04c45eb66a33ec7874e27670fe00473)

   Evt    Hits       MD       LS      T3       T5       pLS       pT5      pT3      TC       Reset    Event     Short             Rate
   avg      2.1      1.4      0.5      2.5      2.9      0.6      1.4      1.1      1.6      0.4      14.7      12.0+/-  2.6      16.1   explicit_cache[s=1]
   avg      3.2      1.7      0.7      3.8      3.8      0.7      2.2      1.6      2.3      0.7      20.8      16.9+/-  3.7      11.2   explicit_cache[s=2]
   avg      6.2      2.3      1.4      6.3      6.2      0.9      4.3      2.8      3.9      1.1      35.4      28.4+/-  5.3       9.3   explicit_cache[s=4]
   avg     10.0      3.6      2.2      8.4      8.9      1.1      6.9      3.8      5.2      1.4      51.7      40.5+/-  8.2       9.0   explicit_cache[s=6]
   avg     15.3      4.2      2.5     10.2     10.7      1.3      8.6      5.1      6.5      1.5      66.1      49.5+/- 11.5       8.6   explicit_cache[s=8]

After (4d2963adaa2db1d226c4efe01be5d217f5106eea)

   Evt    Hits       MD       LS      T3       T5       pLS       pT5      pT3      TC       Reset    Event     Short             Rate
   avg      2.0      1.4      0.5      2.6      2.9      0.6      1.4      1.1      1.6      0.4      14.6      12.0+/-  2.6      15.9   explicit_cache[s=1]
   avg      3.2      1.8      0.8      3.9      3.9      0.7      2.3      1.7      2.4      0.8      21.3      17.4+/-  4.0      11.5   explicit_cache[s=2]
   avg      6.3      2.4      1.4      6.1      6.1      0.9      4.4      2.7      3.7      1.3      35.2      28.0+/-  5.7       9.4   explicit_cache[s=4]
   avg     12.7      3.4      2.0      7.8      8.4      1.1      6.2      3.7      5.4      1.7      52.5      38.7+/- 11.8       9.3   explicit_cache[s=6]
   avg     15.2      4.4      2.7     10.7     11.2      1.5      9.1      5.3      6.5      1.7      68.2      51.5+/- 12.2       8.8   explicit_cache[s=8]

After (4d2963adaa2db1d226c4efe01be5d217f5106eea) - No pLS duplicate cleaning

   Evt    Hits       MD       LS      T3       T5       pLS       pT5      pT3      TC       Reset    Event     Short             Rate
   avg      2.0      1.4      0.5      2.5      2.9      0.0      9.1      1.4      1.6      0.5      21.9      19.9+/-  4.9      23.3   explicit_cache[s=1]
   avg      3.3      1.8      0.9      3.9      4.2      0.0     11.8      2.6      3.2      0.7      32.4      29.1+/-  7.6      16.9   explicit_cache[s=2]
   avg      7.1      2.8      2.1      6.1      6.2      0.0     15.9      4.3      6.6      1.0      52.2      45.0+/- 11.7      13.6   explicit_cache[s=4]
   avg     12.2      3.6      2.5      8.6      9.7      0.0     22.1      6.4      8.0      1.4      74.5      62.3+/- 17.7      13.3   explicit_cache[s=6]
   avg     19.5      5.3      3.4     11.2     12.8      0.0     24.8      8.7     10.7      1.8      98.2      78.7+/- 22.8      13.1   explicit_cache[s=8]

Increase in timing anticipated by the increase of duplicates (seen below).

Physics validation

Before (48e470e0b04c45eb66a33ec7874e27670fe00473) - After (4d2963adaa2db1d226c4efe01be5d217f5106eea) https://uaf-10.t2.ucsd.edu/~evourlio/SDL/PR334/comparePR334_48e470-PU200_48e470D-PU200/

No performance changes

Before (48e470e0b04c45eb66a33ec7874e27670fe00473) - After (4d2963adaa2db1d226c4efe01be5d217f5106eea) - No pLS duplicate cleaning https://uaf-10.t2.ucsd.edu/~evourlio/SDL/PR334/comparePR334NopLSDupClean_48e470-PU200_48e470D-PU200/

Efficiency seems ~unchanged:
Fake rate largely decreased (I don't fully understand it):
Duplicate rate dramatically increased (as expected, when using the legacy pixel tracks as inputs, as we do currently):
In general, we seem to be exchanging fake rate for duplicate rate. Again, this not meant to be configuration to work with for now.

SegmentLinking / TrackLooper

Toggle to disable the pLS duplicate cleaning #335

Timing

Physics validation