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/components/PlaylistsRail.tsx

@@ -70,7 +70,7 @@ export default function PlaylistsRail({
     { key: "name", label: t("playlists.railSortName"), defaultDir: "asc" },
     { key: "count", label: t("playlists.railSortCount"), defaultDir: "desc" },
     { key: "duration", label: t("playlists.railSortDuration"), defaultDir: "desc" },
-    // E10 S2 — "the one I just touched, first". Reads `updated_at`, which the API bumps for item
+    // "the one I just touched, first". Reads `updated_at`, which the API bumps for item
     // adds/removes/moves/reorders too, not just renames (see the backend's _touch). NOT the
     // default: a rail that reshuffles itself while you look at it costs more than it saves.
     { key: "recent", label: t("playlists.railSortRecent"), defaultDir: "desc" },
@@ -178,9 +178,9 @@ export default function PlaylistsRail({
   });
 
   // Enter and Escape unmount the focused <input> and re-render the row as a NEW <button>, so without
-  // this a keyboard user was dropped on <body> and lost their place in the rail (E10 review round 1).
-  // The row is focused from its ref callback, i.e. as that button attaches: measured in Chrome (E11
-  // S2), a focus issued after the commit sticks and nothing takes it back — the round-1 attempt that
+  // this a keyboard user was dropped on <body> and lost their place in the rail.
+  // The row is focused from its ref callback, i.e. as that button attaches: measured in Chrome, a
+  // focus issued after the commit sticks and nothing takes it back — the first attempt that
   // "never worked" focused before the replacement button existed. Set by Enter/Escape only: a
   // blur-commit means the user put the focus somewhere else on purpose.
   const refocusRowRef = useRef<number | null>(null);
@@ -189,7 +189,7 @@ export default function PlaylistsRail({
   // Opening this field from the context menu used to race the menu's own focus return (floating-ui
   // hands the focus back to the trigger from a microtask queued as the popover unmounts, and the field
   // mounted in that same commit). `MenuItemList` now runs the item's action on a fresh task AFTER that
-  // return, so the field simply mounts and is focused here; measured in Chrome (E11 S2), the field is
+  // return, so the field simply mounts and is focused here; measured in Chrome, the field is
   // focused and survives.
   useEffect(() => {
     if (renamingId !== null) renameInputRef.current?.focus();