Fix face detection bbox

[RyoWakabayashi]

bbox is not a reversal of result, but uses all but the last one

https://github.com/opencv/opencv_zoo/blob/main/models/face_recognition_sface/demo.py#L80

result = recognizer.match(img1, face1[0][:-1], img2, face2[0][:-1])

This [:-1] is not "reversed" but "other than last"

[cocoa-xu]

Good catch! I must have read it as [::-1]. Thank you very much for the PR!