fix(design): claim audit part 1b — whole-row switches, the field-family edge, state-layer hovers, icon links, reason-disabled controls, 24/44 targets (E28)

- F12 (owner: A): SwitchRow, the whole settings row toggles; a Config boolean toggles from its title
  and hint (a label pointed at the switch).
- F14 (owner: B): the form dropdown trigger wears the field frame (see-through, the >= 3:1 edge);
  labelled pills keep the faint edge and the accent-edge hover. DESIGN says so, with log rows.
- F6: ~15 bg-card / bg-border / opacity hovers onto the B3 state layer; a lint rule (HOVER_SWAP)
  refuses the swap, contract-tested.
- F10: bare 14px icon links onto IconLink / LinkButton; a local lint rule refuses a glyph-only <a>.
- F13: reason-disabled controls are aria-disabled (Users, Channels backfill, Discovery, Scheduler
  runs, Config tests); FocusSafeButton now swallows the click of a caller's aria-disabled; the
  Config tests say why while dirty (testDirty, HU/EN).
- F5/F11: every control measured live at 24 (mouse) and 44 (coarse) across every page; touch-target
  is a utility that wins over a caller's min-h-6; link variants carry it too; SegmentedToggle,
  Tabs, NavRail, menu rows, PriorityStepper, sliders, CheckRow, InnerInput, card links and the rest.
- F9: DismissibleChip's label reaches the chip's 24; the Combobox chips are DismissibleChips.
- F15: the "+N" / "More..." open state worded as it renders.

frontend/src/index.css

@@ -81,6 +81,16 @@
     background-color: var(--press);
   }
 }
+/* A3: the coarse-pointer target (the why is at the E10 S1c note below). A caller's own `min-h-6` /
+   `min-w-6` (the mouse's 24 floor, B1) is a utility too and sorted after this one, so every control
+   that carried both stayed 24 on a tablet (E28 audit, measured): on a coarse pointer the target wins,
+   hence `!important`. (It would lower a caller's LARGER minimum; no control pairs it with one.) */
+@utility touch-target {
+  @media (pointer: coarse) {
+    min-width: var(--touch-target) !important;
+    min-height: var(--touch-target) !important;
+  }
+}
 /* B3: a filled control mixes its own text colour into its fill (14% hover, 24% pressed). */
 @utility state-filled {
   background-color: var(--accent);
@@ -636,13 +646,8 @@ body {
      button — a worse failure than a small target. Real space cannot overlap.
 
      2.75rem, not 44px: the whole app is rem-derived off `--font-scale`, so a target defined in px
-     would be the one control that ignores the user's text-size setting. */
-  @media (pointer: coarse) {
-    .touch-target {
-      min-width: var(--touch-target);
-      min-height: var(--touch-target);
-    }
-  }
+     would be the one control that ignores the user's text-size setting.
+     (The rule itself is the `touch-target` utility near the top of this file.) */
 
   /* ===== E10 S2 · rows that answer a right-click / long-press (useContextMenu) =====
      The long-press path competes with two browser behaviours that fire at the same gesture: