Add PTZ calibration app and calibrated move parameters

[MateoLostanlen]

Summary

• Add tools/ptz_calibration_app.py — Streamlit app for calibrating PTZ motor speed tables with automated ORB keypoint matching
• Recalibrate all speed/bias tables for both 823A16 and 823S2 (pan + tilt) using server-side timing
• Add zoom-aware speed selection: use full speed table at zoom 0, speed 1 only when zoomed in
• Add micro-pulse support: zoom to 41 then move+stop for precise small-angle nudges

Key changes

routes_control.py

• Updated PAN_SPEEDS, PAN_BIAS, TILT_SPEEDS, TILT_BIAS for both camera models
• _pick_speed() now accepts zoom param — restricts to speed 1 when zoom > 0
• Extracted _execute_axis() helper in click_to_move (removes pan/tilt duplication)
• Micro-pulse zooms to 41 before firing for consistent displacement (~1.6° pan, ~2.1° tilt on 823A16)
• Skip threshold changed from hardcoded 0.5° to bias(speed1) / 2
• Added /control/speed_tables endpoint for external tools
• Server-side duration parameter: move+sleep+stop in single API call (no VPN latency)

tools/ptz_calibration_app.py

• Automated calibration via ORB keypoint matching (2000 features, Lowe's ratio test)
• Zoom sweep mode to measure speed vs zoom across multiple levels
• Manual annotation fallback for micro-pulse at high zoom
• Click-to-move accuracy test tab
• Fetches speed tables from API to avoid duplication
• Incremental JSON save during long sweeps

tools/ptz_zoom_speed_calibration_report.md — Research data showing Reolink internally caps PTZ speed when zoomed in

tools/README.md — Step-by-step calibration guide

Context

Previous speed tables were calibrated through VPN with separate move/stop API calls, adding 50-200ms uncontrolled latency to each measurement. Server-side timing eliminates
this. A zoom sweep revealed Reolink cameras limit motor speed at zoom > 0 (all speeds cap to ~1.5°/s), so higher speeds are only useful at zoom 0.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 72.59%. Comparing base (f8b6e05) to head (a07e09f).
⚠️ Report is 3 commits behind head on develop.

Additional details and impacted files

@@             Coverage Diff             @@
##           develop     #335      +/-   ##
===========================================
- Coverage    79.97%   72.59%   -7.39%     
===========================================
  Files            7        7              
  Lines          824      602     -222     
===========================================
- Hits           659      437     -222     
  Misses         165      165              

Flag	Coverage Δ
unittests	72.59% <ø> (-7.39%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.