BforSpectroscopy optimise

To fix bugs and optimise Hlt2 selections in BforSpectroscopy: https://gitlab.cern.ch/lhcb/Moore/-/blob/master/Hlt/Hlt2Conf/python/Hlt2Conf/lines/bandq/builders/b_for_spectroscopy.py?ref_type=heads

Bug fixed:

• fix wrong pi charge in B+ -> D- pi+ pi+ decay descriptor
• fix make_dp -> make_omegac function in tight_Omegac
• enlarge mass windows for jpsiX channels, from +-80MeV (too tight) to [-120;+220]MeV.

Selections optimized:

• increase cut on B bpvltime_min from 0.2ps to 0.4ps, which corresponds to ctau>0.12mm (tau_B ~ 1.5ps)`, keep 0.3ps for JpsiX modes and 0.15ps for Bc->JpsiX modes;
• add b_bpvdirac_min=0.995 (tan<0.1), 0.95 for SL (tan<0.33) and charm_bpvdira_min=0.98 (tan<0.2) - justified by RapidSim;
• loosen bpvvdchi2_min from 36 to 25
• add +100MeV for upper mass sideband (suggested by Steve)

Expected to give up to 3-4x reduction in rate.