Resolve "Make minMarkerPerimeter and maxMarkerPerimeter user-configured for Multi+Aruco"

This makes the minMarkerPerimeter and the maxMarkerPerimeter changeable by the user even when using the Multicolormarker with Code instead of the Codemarker.

The MarkerPerimeter property is a multiplier. If the image-patch would be 640x480 and minMarkerPerimeter would be 0.1, the mininal perimeter would be 0.1 * 640 = 6.4 approx. 6 Pixels. Since we set the perimeter in cm, the corresponding number of pixels is calculated (cm per pixel is different for different parts of the image in slanted view, we just use the same method used in the global code marker).

This slightly changes the results of the regression for the TdN test(4 frames). The other Multi+Code test needed to be adjusted, but was only changed in the MarkerPerimeter, whose interpretation changed with this MR.

___ TestTrackerTxt.test_e2e_via_txt[multiColorMarkerWithAruco_dictMip36h12] ___
Diverging coordinates at Frame: 318 X: 1.04125 Y: -0.0725276 Z: 1.0 in the test file.
At Frame: 318 X: 1.0528 Y: -0.0995424 Z: 1.0 in accepted truth.
For person ID: 1 in test file and person  ID: 1 in truth file

Diverging coordinates at Frame: 319 X: 1.06121 Y: -0.0930144 Z: 1.0 in the test file.
At Frame: 319 X: 1.07199 Y: -0.120554 Z: 1.0 in accepted truth.
For person ID: 1 in test file and person  ID: 1 in truth file

Diverging coordinates at Frame: 320 X: 1.08012 Y: -0.114633 Z: 1.0 in the test file.
At Frame: 320 X: 1.08994 Y: -0.14115 Z: 1.0 in accepted truth.
For person ID: 1 in test file and person  ID: 1 in truth file

Diverging coordinates at Frame: 321 X: 1.10015 Y: -0.136449 Z: 1.0 in the test file.
At Frame: 321 X: 1.10877 Y: -0.163541 Z: 1.0 in accepted truth.
For person ID: 1 in test file and person  ID: 1 in truth file

(Tip: For running it locally, so you have the test data, you can use sth. like: python -m pytest --path="/path/to/petrack.exe" -k [multiColorMarkerWithAruco_dictMip36h12] from the tests dir)

Closes #210 (closed)