ADR-0024 Phase F (full) step 4: catalog token-keyword cleanup

Drops the 47 `parse.token.keyword.*` and 6 `parse.token.punct.*`
catalog entries (and their `KEYS_AND_PLACEHOLDERS` declarations).
Nothing consumes them: the walker renders keyword wording in
`format!(\"`{word}`\")` directly, sourced from grammar-tree Word
literals; punct wording surfaces the same way via
`Expectation::Punct(ch)`.

Structural-class labels (`parse.token.identifier`,
`parse.token.number`, `parse.token.string_literal`,
`parse.token.flag`, `parse.token.end_of_input`) and the lex-error
wordings (`parse.token.error.{bad_flag,unknown_char,
unterminated_string}`) stay. These are not derivable from the
grammar tree and the walker's expected-set / validator paths still
read them.

`friendly::keys::tests::keys_validate_against_catalog` continues to
assert catalog ↔ `KEYS_AND_PLACEHOLDERS` bidirectional coverage,
so the trimmed declaration is pinned against the trimmed catalog.

Tests: 806 passing, 0 failing, 1 ignored. Clippy clean.

src/friendly/strings/en-US.yaml

@@ -372,67 +372,16 @@ parse:
     import: "import <zip-path> [as <target>]"
     mode: "mode simple  |  mode advanced"
     messages: "messages  |  messages short  |  messages verbose"
-  # Single-token vocabulary the renderer uses to translate
-  # chumsky's expected-set patterns. One key per Keyword variant
-  # (validated against `Keyword::ALL`), one per Punct variant,
-  # one per token-class label, one per LexError kind.
+  # Single-token vocabulary (ADR-0021 §4, ADR-0024 §cleanup-pass).
+  # The per-keyword and per-punct entries collapsed when the
+  # walker became the source of truth — keyword wording is now
+  # produced verbatim as `` `<word>` `` by the parse-error
+  # formatter, sourced from the grammar tree's Word literals.
+  # What remains here is structural-class labels (identifier /
+  # number / string literal / flag / end of input) and the
+  # lex-error wordings, none of which are derivable from the
+  # grammar.
   token:
-    keyword:
-      create: "`create`"
-      drop: "`drop`"
-      add: "`add`"
-      rename: "`rename`"
-      change: "`change`"
-      show: "`show`"
-      insert: "`insert`"
-      update: "`update`"
-      delete: "`delete`"
-      replay: "`replay`"
-      table: "`table`"
-      column: "`column`"
-      data: "`data`"
-      relationship: "`relationship`"
-      pk: "`pk`"
-      with: "`with`"
-      from: "`from`"
-      to: "`to`"
-      into: "`into`"
-      as: "`as`"
-      in: "`in`"
-      on: "`on`"
-      set: "`set`"
-      where: "`where`"
-      values: "`values`"
-      "null": "`null`"
-      "true": "`true`"
-      "false": "`false`"
-      cascade: "`cascade`"
-      restrict: "`restrict`"
-      action: "`action`"
-      "no": "`no`"
-      # App-lifecycle commands (per ADR-0003, surfaced through
-      # the parser to drive completion + usage templates).
-      quit: "`quit`"
-      help: "`help`"
-      rebuild: "`rebuild`"
-      save: "`save`"
-      new: "`new`"
-      load: "`load`"
-      export: "`export`"
-      import: "`import`"
-      mode: "`mode`"
-      messages: "`messages`"
-      simple: "`simple`"
-      advanced: "`advanced`"
-      short: "`short`"
-      verbose: "`verbose`"
-    punct:
-      colon: "`:`"
-      open_paren: "`(`"
-      close_paren: "`)`"
-      comma: "`,`"
-      equals: "`=`"
-      dot: "`.`"
     identifier: "identifier"
     number: "number"
     string_literal: "string literal"