when looping, if looking for previous_dbname, would always be defined because of the last loop, so comparing the wrong databases

default value is core, so running without specifying anything will be as before

when the script is run, adds new entries for each seq_region id which has changed between current and previous release
but also updated existing entries so that all rows internal_seq_region_id in seq_region_mapping correspond to a seq_region_id which is in the current database

also supports trinomial species names

for coord_system, check rank and name

methods are cleaner
added length check to decide if it is the same region

password is not always required

changes to get it to do what I think it should do - 1) staging and staging2, 2) add an initial mapping_set entry if this is  new species 3) warn if there is no mapping_set entry but this is not a new species - in this case I the genebuilders should be adding the old one since this is the scenario we want to catch! (need to check the assembly hasn't changed), 4) dry_run, ie preview mode

many changes but the most important is to only add seq_region mapping if the seq_region_ids are different