M12f: remove the 19 string keys R8 found unused (20 CSV rows); §9.11 tie-break comment

Removed from the sv/en resources and docs/ux/strings.csv (562 rows, header included): common_archived_category,
editor_amount_hint, editor_split_half_each, editor_split_preview_effect, editor_next, analytics_range_dates,
analytics_chart_spoken, analytics_axis_thousands, expense_detail_title, expense_detail_account,
home_notification_rationale, home_sync_status_all_synced_body, home_sync_status_tag_new, home_sync_status_tag_changed,
home_sync_status_tag_deleted, home_sync_status_row_pending, settings_devices_summary (one, other),
settings_pending_badge, settle_empty_settled. CategorySuggestion cites §9.11 instead of a stale TODO(spec).

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01TQd3pEP9UeN2uifbHknQ7B

android/domain/src/main/kotlin/se/hagstedt/kvitt/domain/CategorySuggestion.kt

@@ -69,7 +69,7 @@ fun normalizeDescription(text: String, unicode: UnicodeText): String {
  * CS-5: history exact, then history first token, then the keyword table (CS-6), then [fallback] (the AC-8.6 default).
  * An empty normalised description gives [fallback]. History counts only categories in [activeCategories]; the most
  * often used wins, a tie goes to the category whose latest expense is latest (`expenseDate`, then `createdAt`), and a
- * remaining tie to the lower category id (TODO(spec): not stated in category-suggestion.md).
+ * remaining tie to the lower category id in string order (SPEC §9.11, "Most recently").
  */
 fun suggestCategory(
     description: String,