Restore sequential hero copy and first-appearance entry

data/tv-spec/spec.md

@@ -66,16 +66,16 @@ Home heroes and shelves, Library rows, Discover hero/lanes, title headers, episo
 | Poster caption in / out | 60 ms delay + 180 ms ease-out / 100 ms ease-out | 10 px rise on entry only; exit holds its position, including when entry was interrupted |
 | Network artwork only | 160 ms, Coil painter alpha | None; memory, disk and local hits are instant |
 | Row/tab content reveal | 180 ms, ease-out | One layer per row, 24 px rise |
-| Initial title copy | 180 ms, ease-out | 16 px rise; already-loaded Home/Discover hero copy paints immediately |
+| Initial hero/title copy | 180 ms, ease-out | 16 px rise (`copyEntryRise`); saved hero entry state suppresses replay on remount, not first appearance |
 | Hero backdrop hand-off | 320 ms, ease-in-out (`.65,0,.35,1`) | Alpha only; retain previous art until incoming art or its final fallback is ready |
-| Hero copy hand-off | Incoming and outgoing alpha counter-fade over 200 ms ease-out, after artwork or final fallback is ready | 14 px rise for incoming copy only; interrupted outgoing copy holds its current translation |
+| Hero copy hand-off | Outgoing 120 ms ease (`.25,.1,.25,1`); incoming starts after 140 ms and enters over 200 ms ease-out, gated on artwork or final fallback readiness | One copy slot: no overlapping text; 14 px incoming rise, outgoing translation held where entry stopped |
 | Late ratings, genres, overview and readiness | 200 ms, ease | Alpha only, inside existing measured slots; rating plates keep their existing plate fade rather than adding a nested fade |
 | Horizontal / vertical focus-follow scroll | 160 ms ease / 180 ms ease-out | Existing scroll geometry |
 | Discover pager | 180 ms, ease-out, retargeting current width and opacity | Buttons and pager remain outside the copy transition; reduced motion snaps width and caps opacity fades at 80 ms |
 
 Row reveals are grouped rather than the browser prototype's per-card staging: the armv7 frame budget permits one animated layer per shelf, not one per card. Tabs have no full-screen transition layer. The upper Home hero blend is bounded to **1920×820 px**, Discover to **1920×740 px**; the entire painted backdrop stays static while incoming artwork loads, and the lower backdrop changes only when artwork or its final fallback is ready, without an animated screen-sized layer. Hero copy is a separate bounded block with stationary action controls. Ordinary image reveal uses Coil's painter; hero-controlled requests bypass that independent fade so only the token-backed hero blend owns the transition. Existing poster-only blur is static.
 
-The title page's scroll dim is a draw-only colour overlay, not an alpha-animated full-screen layer. Discover scrolls its backdrop through placement offsets rather than a full-screen translation layer. Initial loading animated layers are Home **2**, Library **up to 3** lazy rows, Discover **2**, title header **1**, person **2**; pending title episode/cast shelves add **1 each**, and each pending Discover lane owns **1**. Deep Home loading is bounded to **4** visible shelf layers. These are animated layers, not independent pulse clocks or every static display-list node: one shared pulse clock drives the loading screen without adding a layer. An ordinary hero hand-off animates **3** bounded owners (outgoing art and two copy blocks) above an opaque artwork base. Interruptions retain the painted composite without resetting alpha; each outgoing owner keeps its own original 320 ms artwork / 200 ms copy lifetime rather than restarting with later presses. Focus changes can animate **2** card transforms, **1** caption and at most **2** rating plates; ordinary artwork fades are painter draws without additional compositing layers. Late title metadata adds at most **4** bounded alpha layers: title ratings, play-target ratings, genres and overview. No animated owner is full-screen.
+The title page's scroll dim is a draw-only colour overlay, not an alpha-animated full-screen layer. Discover scrolls its backdrop through placement offsets rather than a full-screen translation layer. Initial loading animated layers are Home **2**, Library **up to 3** lazy rows, Discover **2**, title header **1**, person **2**; pending title episode/cast shelves add **1 each**, and each pending Discover lane owns **1**. Deep Home loading is bounded to **4** visible shelf layers. These are animated layers, not independent pulse clocks or every static display-list node: one shared pulse clock drives the loading screen without adding a layer. An ordinary hero hand-off animates **2** bounded owners (outgoing art and one sequential copy block) above an opaque artwork base. Art interruptions retain the painted composite without resetting alpha; each outgoing art owner keeps its original 320 ms lifetime. Copy retargets its current alpha, fades out before replacing text, then uses the approved delay/entry tokens; never two copy layers are visible together. Focus changes can animate **2** card transforms, **1** caption and at most **2** rating plates; ordinary artwork fades are painter draws without additional compositing layers. Late title metadata adds at most **4** bounded alpha layers: title ratings, play-target ratings, genres and overview. No animated owner is full-screen.
 
 The approved title follow-up (2026-11-07) supersedes the fixed readiness reservation: an absent readiness line occupies **0 px**, while a present line occupies **56 px (28 dp)**. Late arrival/removal animates that height over the **200 ms late-line duration with token ease-out**; reduced motion snaps. Overview, actions and following rows move monotonically with that band; an already-pinned focused row stays pinned through same-remeasure scroll compensation. Ratings remain inside the fixed-height facts band and appear once: an explicit episode page uses its opened episode's own set, including the episode score, rather than the series-only set. Series and episode pages have no redundant Seasons shortcut; Down reaches the season chips. Skeletons have no click/hover focus targets; pending episode/cast rows remain absent from the key-navigation state until real cards exist. Initial title/person loading retains destination focus rather than focusing the root nav; Down cannot enter a skeleton, and Back still pops the route without opening the picker. An already-ready catalogue or held title stays Ready while a same-route, same-person refresh reads new data; the existing signed-in authentication guard remains authoritative, and loaded copy/rows do not remount or replay their entry motion.
 
@@ -85,7 +85,7 @@ Discover key traversal skips pending lanes in both directions. If every remainin
 
 Native **1920×1080** before/after frames and the frame-budget inventory are published at [the motion build proof](https://pages.justinvisser.org/wolfskin/tv-motion-build/). Before frames use the pre-change surface sources in this same isolated worktree; after frames use the shipped Compose surfaces. No device access, installation or optimized APK is part of this lane. On-box acceptance remains: no series layout change after first content, Home scrolling no worse than **39.6% janky / p90 65 ms**. Native JVM rendering does not measure that GPU/frame-timing target.
 
-The [round-2 Discover native proof](https://pages.justinvisser.org/wolfskin/tv-round2-discover/) publishes paired 1920×1080 frame strips and exact 16 ms alpha/translation/position series for ready, slow, failed, sustained rapid-selection and reduced-motion scenarios. Both versions use identical fixture data, immediate key-down/up, and the same selection timestamps. The regression suite checks an opaque artwork base, visible copy composite, monotonic per-layer transitions, stationary controls and token-bounded outgoing lifetimes. These are native-rendering proofs, not on-box frame-pacing measurements.
+The [round-2 Discover native proof](https://pages.justinvisser.org/wolfskin/tv-round2-discover/) publishes paired 1920×1080 frame strips and exact 16 ms alpha/translation/position series for ready, slow, failed, sustained rapid-selection and reduced-motion scenarios. The reviewed counter-fade remains available separately for comparison with the corrected sequential copy; initial-entry/remount and Home hero series are also published. The regression suite checks an opaque artwork base, at most one visible copy, monotonic per-phase motion, stationary controls and token-bounded outgoing artwork lifetimes. The approved fade-through may briefly have no visible copy; it never paints overlapping titles. These are native-rendering proofs, not on-box frame-pacing measurements.
 
 
 ## Card metadata and bottom state line — revision 6, option B