iris APK only. Three corrections to the report, two of them shipped this
morning: `missed vsyncs` counted the bench's own idle pauses as stutter,
`late` counted the wait for the display as cost, and the display reported
60Hz for a run that drew at 120 -- so every late figure was judged
against twice its budget.
Also corrects the frame-rate reading in the previous entry: both runs
sustained ~120.3fps through the motion, so the loop was never dropping
frames and the clock is what made it smooth.
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>