Minor UI bug fix with the altitude display.

Background refreshes have been tuned for a rock-steady 15-minute cadence — your pressure trace now flows uninterrupted from dawn to dusk, with no mysterious gaps where the barometer went quiet. Storm alerts are yours to shape. Dial sensitivity to Severe for only the serious systems, Moderate for the meaningful shifts, or All to catch every tremor in the atmosphere. Pair it with a daily cap and Pascal will respect your attention — a quiet companion when the sky is calm, a sharper voice when the pressure starts to fall.

Bug Fixes: Fixed: Rapid pressure drops (storms) could permanently block new readings Fixed: Race condition causing missed samples on launch and in background Improved outlier detection that handles extreme weather without false rejections

Weather Forecasting - Improved prediction accuracy and reliability. Altitude Mode - You can now display your altitude over-time - Data points visible on altitude chart for easier reading - Chart toggle moved below the forecast card for cleaner layout Chart Fixes - Fixed axis labels getting cut off on pressure and altitude charts - Improved chart padding for better readability

### Added - **On-Device Machine Learning**: Personalized weather predictions that learn automatically - **Instant Bootstrapping**: Uses existing pressure history to personalize immediately - **Adaptive Forecasting**: Combines 110-year-old Zambretti algorithm with modern ML - **Overnight Sampling**: Charging-aware background scheduling for better overnight coverage - **Diurnal Correction**: Removes natural daily pressure cycle for more accurate trend detection - **Storm Alerts**: Push notifications for rapid pressure drops and storm conditions - **Live Activities**: Dynamic Island and Lock Screen pressure display (iOS 16.1+) - **Flight Mode**: Cabin pressure tracking with altitude estimation ### Fixed - Background processing tasks now run properly overnight when charging ### How ML Works Pascal learns your local weather patterns automatically — no user action required: 1. **Instant bootstrap** — Seeds model with Zambretti archetypes on first launch 2. **History training** — Walks through existing pressure data to create observations 3. **Auto-evaluation** — After 6 hours, Pascal checks what actually happened 4. **Kernel likelihood** — Similar past conditions inform future predictions **Result:** If you have a week of pressure history, Pascal is personalized immediately. New users reach personalization after ~25 auto-evaluated predictions. All learning happens silently in the background. No feedback buttons needed.