Bug fix for the Upsilon trigger high mass window

This is a bug fix MR.

The dimuon builder for upsilon is taken from charmonium_to_dimuon, see https://gitlab.cern.ch/lhcb/Moore/-/blob/master/Hlt/Hlt2Conf/python/Hlt2Conf/lines/bandq/builders/dimuon_lines.py?ref_type=heads#L26

The baseline builder set a higher mass limit of 10GeV, see https://gitlab.cern.ch/lhcb/Moore/-/blob/master/Hlt/Hlt2Conf/python/Hlt2Conf/lines/charmonium_to_dimuon.py?ref_type=heads#L176

This is not overwritten or removed in later builders. 10GeV kills Upsilon2S, Upsilon3S.

Noticed when testing lhcb-datapkg/AnalysisProductions!1103 (diffs) .

May influence a few lines but let's try if we can fix Upsilon first quickly to be included in the next round of data-taking.

As a quick fix I change Upsilon line mass upper limit to 12GeV to cover the highest bbbar in PDG: https://pdglive.lbl.gov/ParticleGroup.action?init=0&node=MXXX030 , and leave some space for new states search.

Local test: no impact on rate

Upsilon3S efficiency after the update: 27%

logs/Upsilon3S_new.log:    LAZY_AND: Hlt2BandQ_DiMuonUpsilonFullDecisionWithOutput                                                             #=100     Sum=27          Eff=|( 27.00000 +- 4.43959 )%|

Upsilon3S efficiency before the update: 0%

logs/Upsilon3S_old.log:    LAZY_AND: Hlt2BandQ_DiMuonUpsilonFullDecisionWithOutput                                                             #=100     Sum=0           Eff=|( 0.000000 +- 0.00000 )%|