chore(comments): drop epic, sprint and review-round ids from the code

The code cited its plan instead of saying why: `E28 review J4`, `(R6 S3b)`, `S1 justified…`,
`deep-review round 5`. That plan lives in the private ops repo, so for every reader of this one the
citation was noise, and a sentence whose subject was the plan step ("round 2 showed…") hid its
lesson behind an id. Each one is gone and the "why" kept: a bare citation is deleted, a sentence
carrying a lesson is reworded around it ("a later attempt showed…").

Measured (`scripts/audit_metrics.py`, product code): 652 lines in 270 files carried an epic code by
the definition the next commit holds (507 / 241 by the old, two-digit-only one) -> 0. Across all
code, tests, scripts and configs: ~1100 epic-code lines plus ~320 sprint/round ids.

Behaviour unchanged, checked mechanically: every changed .py file has the same AST once docstrings
are blanked; css/sh/toml/yml/json files are identical once comments are stripped; the only
non-comment changes in TS/JS are test titles, one description string in a contract table, and five
ESLint messages, whose text is all the gates snapshot diff holds. Not touched: Markdown (docs and
the changelog keep their history), Alembic migrations, TV episode notation (`S2 · E5`), DESIGN.md
ids (`C1`, `B4`, `S3h`), tracker ids (`SEC-16`, `DX-17`).

One stale claim fixed on the way: test_hls_reaper's docstring said the player checkpoints "every
10 s whether or not it is playing", which plex/stream.py calls wrong; it now says what the code does.

frontend/src/lib/useFocusKeeper.ts

@@ -16,7 +16,7 @@ const FOCUSABLE =
   'button:not([disabled]), a[href], input:not([disabled]), select:not([disabled]), textarea:not([disabled]), [tabindex]:not([tabindex="-1"])';
 
 // A focusable element the CSS has taken off screen (a `tabletHidden` column's header button) refuses
-// `focus()`, and the focus it was meant to catch fell to <body> (E28 F2, measured in Chrome). An engine
+// `focus()`, and the focus it was meant to catch fell to <body> (measured in Chrome). An engine
 // without `checkVisibility` (jsdom) paints nothing, so there everything counts as shown.
 const shown = (el: HTMLElement) => !el.checkVisibility || el.checkVisibility();
 
@@ -51,7 +51,7 @@ export function focusNeighbour(el: HTMLElement): boolean {
 
 /** Run `remove`, which takes `el` off the page, without dropping `el`'s focus: it goes to the
  *  neighbour first (`focusNeighbour`). That neighbour may go in the same removal — the "Clear all"
- *  that only shows beside two filters, the "+1" of the last hidden one (E28 review round 3) — so
+ *  that only shows beside two filters, the "+1" of the last hidden one — so
  *  once the removal has rendered, a focus left on <body> goes to what preceded `el`, else to the
  *  first control left in its row. Only when the focus was lost: a focus the reader moved stays. */
 export function handOffFocus(el: HTMLElement, remove: () => void): void {
@@ -90,7 +90,7 @@ export const FOCUS_REST_ATTRS = { "data-focus-rest": "" } as const;
 /** Whether the focus is nowhere the reader chose — <body>, the keeper's stand-in, or a resting place
  *  (`FOCUS_REST_ATTRS`) — so a restore may take it. The page region used to count as a choice: a
  *  click on its background focused it, and Back then left the focus there instead of on the item
- *  the view was opened from (E28 review H1). */
+ *  the view was opened from. */
 export function focusIsUnclaimed(el: Element | null = document.activeElement): boolean {
   return !el || el === document.body || isStandInFocus(el) || el.hasAttribute("data-focus-rest");
 }