fix CommonParticles for DaVinci tests

fix CommonParticles for DaVinciTests

• (probably only partial) fix for new signatures for some LoKi-functors
• sed-replace of BPVDIRA with BPVDIRA()
• sed-replace of BPVVDCHI2 with BPVVDCHI2()
• sed-replace of VFASPF(VCHI2) with CHI2VX
• sed-replace of VFASPF(VCHI2/VDOF) with CHI2VXNDF
• sed-replace of ....
• StdLooseDplus.py :
  • make use of in_range
  • change order of functors
  • remove part of redundand parentheses
  • make code more readable
  • ...

btw, for @gligorov (the author of StdLooseDplus.py) : one can easily make it more efficient

• using DaVinci::N3Body algorithm instead of CombineParticles - here large gain in CPU is possible
• changing order of cuts
• using ASUM(PT) instead of explicit sum over 1,2,3 particles
• replacing decay descriptor, moving more rare particles to the start, e.g. K pi pi is for sure more efficient (especially for DaVinci::N3Body, for CombineParticles the gain is probably small, since the algorithm itself is slower) than pi pi K