fix(detection): scale from_tensorflow boxes by correct axes

[RubenHaisma]

Description

Detections.from_tensorflow scaled the normalized box coordinates by the wrong image dimensions. Tensorflow Hub object-detection models emit detection_boxes as normalized [ymin, xmin, ymax, xmax] (the connector's own [1, 0, 3, 2] reorder confirms this layout), so the y coordinates must scale by height and x by width. The code did the opposite:

boxes[:, [0, 2]] *= resolution_wh[0]   # ymin, ymax scaled by WIDTH  (should be height)
boxes[:, [1, 3]] *= resolution_wh[1]   # xmin, xmax scaled by HEIGHT (should be width)

The bug is masked on square images (width == height) but corrupts every coordinate on the common non-square case.

Reproduction

A box normalized to [ymin=0.1, xmin=0.2, ymax=0.5, xmax=0.6] on a 1000×500 image (resolution_wh=(1000, 500)):

	xmin	ymin	xmax	ymax
before (wrong)	100	100	300	500
after (correct)	200	50	600	250

Correct values: xmin = 0.2·1000, ymin = 0.1·500, xmax = 0.6·1000, ymax = 0.5·500.

Fix

Swap the two multipliers so y scales by resolution_wh[1] (height) and x by resolution_wh[0] (width). No behavior change on square images.

Tests

Adds test_from_tensorflow_scales_axes_on_non_square_image (the connector was previously untested — no references in tests/ or docs/). It fails on develop ([100,100,300,500]) and passes with this fix ([200,50,600,250]). ruff check/format clean.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 82%. Comparing base (09b2199) to head (3ea9854).

Additional details and impacted files

@@           Coverage Diff           @@
##           develop   #2360   +/-   ##
=======================================
  Coverage       82%     82%           
=======================================
  Files           68      68           
  Lines         9560    9560           
=======================================
+ Hits          7881    7886    +5     
+ Misses        1679    1674    -5     

🚀 New features to boost your workflow:

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