A fix to the dynamic RoI settings in TrigFastTrackFinder
This MR is related to !47835 (merged) and !48052 (merged)
We observe a bug in dynamic RoI test: https://idtrigger-val.web.cern.ch/idtrigger-val/TIDAWeb/TIDAart/?hpage=l2eff&type=FS&jobdir=master/Athena/x86_64-centos7-gcc8-opt/2021-11-09T2101/TrigInDetValidation/test_trigID_all_ttbar_generic_pu55
The fsjet instance got too few tracks in the test, while the electron, muon and tau instances work well. After checking, we find that this bug is related to the fullScan option of the RoI setting as we discussed in !47835 (merged). With the fullScan option set to be true, the test will work.