Release v0.0.12

Added

• Explicit ordering in Explorer! The column view now supports pressing 'a' for
  "ascending" and 'd' for "descending. It also supports left and right arrow for
  cycling through the options. If you want to order on multiple columns,
  those columns must be displayed in the order that you want the directives
  applied. Use '+' and '-' to re-order columns. If order is specified on a
  hidden column, that order is ignored.

• Some schematized fields weren't available for query in CLI or Explore. All fields
  are now available. New fields include is_undelete, last_hash_scan_old
  hash_old, hash_new, last_val_scan_old

• Explore columns now use shortened display names which are acronyms in the case
  of very long field names. For example, last_val_scan_old is LVSO

• Documentation update to include all available schema fields in the "book"

Fixed

• Corrected "not_null" to "not null" in Explorer filter tips

• In Explorer, date filter values are validated before the filter is saved.
  Previously, dates were parsed but not actually validated so values such as
  "2025-01-32" would slip past the parser and into the query

• Fixed panic occurring when the window width was too small to display the
  vertical scrollbar in the grid

Changed

• Cleaned up StringFilter - enum types have had their own filter type for
  a while but StringFilter still had legacy enum validation logic

• Internal cleanup of Explore's "column" data structures. Collapsed
  ColumnOption and ColumnInfo

• Grid rows are no longer cloned prior to drawing the table. The rows
  are re-created (no way to avoid this while using table state) but the
  String contents are no longer cloned