Removed regulatory feature tables.

Added patch identifiers to meta table.

fetch_all_by_feature_class method now throws if there is an orphen analysis_id foreign key in the "feature_class" table

character limit on column 'attrib_type.code'.

they are.

  Type 75 (GeneNo_RNA_pseu):  Missing a tab character.

  Type 131 (vega_export_mode):  Type code changed to 'vega_export_mod'
  to conform to 15 character limit of column attrib_type.code.

from the list of tables to process.

  Remove hard-coded external_db_id 11000.

  Change xref.info_type from 'MISC' to 'COORDINATE_OVERLAP'.

The UCSC external_db_id is now in external_db table, so delete hardcoded
external_db_id '11000'.

Change xref.info_type from 'MISC' to 'COORDINATE_OVERLAP'.

commit.

#
# Title: Add new info_type to xref table
#
# Description:
# Add the ENUM 'COORDINATE_OVERLAP' to the xref.info_type column.

slowest method, about 100 times slower than the most basic 'count'
binning method, but it now only uses 0.5 seconds to bin 2360 repeat
features into 100 bins instead of 6-9 seconds...

release document:  Make sure that the regular expression doesn't match
too much.