Read .itm file contents to create migrations for any database columns that don't exist

[kugra-al]

• Job to read item files in batches to parse the file contents
• Cache fields and values so we don't need to read twice
• Reading job should be it's own job. Creating migrations should be it's own job
• Check existing columns in items table.
  • Create migrations for any columns that don't exist.
  • Create migrations to delete any unused columns that don't contain data (will end up with unneeded columns otherwise if there's a field typo in a .itm file)