docs(e2e): correct 19a8d133db — the Korean preedit spec has been executed

That commit said the landed form had never run and carried the probe's
values as literals. It has now run on real hardware: 1 passed, 9.1s, rc=0,
with the capture and log sealed under a verified hash manifest.

The teeth check was also run rather than reasoned about, and it changes
which assertion matters. Forcing the active overlay to max-width:0 with
overflow:hidden — invisible on screen — leaves the active class, the
textContent, display:block AND checkVisibility() all passing. Only
during.rect.width fails. checkVisibility() is not sufficient against this
defect; the bounding rect is the single load-bearing assertion, which the
docblock already said and this run confirms.

An earlier teeth attempt injected the CSS mid-run and tripped the
hasActiveClass poll instead, failing at the wrong assertion. It is
inconclusive and excluded from the seal rather than counted.

tests/e2e/terminal-macos-2set-korean-preedit-visibility.spec.ts

@@ -30,6 +30,12 @@
  * Values below were recorded from a live run under `com.apple.inputmethod.Korean.2SetKorean`;
  * see `11914-c3-live-preedit-visibility.txt` in the lane capture directory. That capture is
  * gitignored and is cited, never imported.
+ *
+ * Correction to this file's own landing commit (19a8d133db7), which said the spec had not been
+ * executed in this form: it has. It passes on real hardware — 1 passed, 9.1s, rc=0 — and the
+ * teeth check above was run against it rather than reasoned about. Both logs are sealed
+ * alongside the capture. An earlier teeth attempt that injected the CSS mid-run tripped the
+ * `hasActiveClass` poll instead of the rect assertion; it is inconclusive and excluded.
  */
 import { execFileSync } from 'node:child_process'
 import path from 'node:path'