22d7c88cf5
P2 shipped the wrong rule and the title is the counter-example. I had reasoned that the exit is the final untimed keyframe, so anything timed was still the entry. It is not: `pteff02` on the title holds at t=46 with the 25 % dim quad at alpha 0x40 and then ramps to 0x00 by t=236. The exit can be a long run of TIMED keyframes, and running to the end drops the dim and leaves the whole screen ~13/255 too bright -- exactly the luminance excess P2 recorded (oracle 64.1, rest 62.8, timeline 80.0) and filed as an unexplained "glow slab". A group is pre-roll -> ramp in -> hold -> ramp out -> post-roll, and a screen that has arrived sits on the hold. `settle_units()` is now `rest.t`, the decoders' own identification of that hold. The check is that a disagreement vanishes: on ALL TWELVE screens the settled timeline is now byte-identical to the `--pose=rest` render, where before this four differed by up to 247/255. The timeline's endpoint should BE `rest` -- the animation is what it adds, not a different destination -- so this is the property to want, and it holds with no special case. Credit where due: this came out of the RE agent measuring the title's dim quad against a plate-free capture, in pursuit of a different question.