Update to description line parsing by EG's request. Seaprate multiple description lines with ; rather than concatenating them together into an un-fathomable string.

1. Fix check_for_stops to cope with multiple adjacent internal stops without a loss of codon to AA sync.
Found when running add_selcys.pl on zebrafish for Vega 47.

2. Fix to existing conditional in vega Sec QC script which tests for an anacode bug before deciding to complain
about missing Sec markup for selenoproteins that also stop with TGA, so that this conditional actually spots
this case in more proteins and so cries wolf less often.

sql to get xrefs with failed priority produces duplicate xref_ids, added IGNORE to INSERT statement for xref

Adding a check in here to highlight when a regular expression failed to match meaning we've found a corrupted schema patch identifier which should be removed

will use 0 instead

the BaseParser expects error_codes and no fatal errors to work around checking for existence of xrefs,
so I turned them off again

unique genes method (for assembly patches), interactive does have an argument, other binaries default to 1 if nothing specified

* reenabled text SQL error messages, as error codes make it impossible to find out which error happened in which statement