Small improvement for sample stereo_match.cpp (Patch #1936)

I think it make sense that when scale != 1 we also multiply the calibration matrix by scale. In the code the images are resized according to scale but then the calibration matrix M1 and M2 are not transformed accordingly.
So in my version I just added the following

1if( scale != 1.f )
2{
3    M1 = scale * M1;
4    M2 = scale * M2;
5}

at around line 193 (anyway after the matrices are read from the calibration file).
With that I get smoother results that are more similar to the case scale = 1.0.
I don't know the details of the dense stereo matching algorithms but this make sense in theory: if u are using, eg, a downsample image then also the calibration matrix is affected.

Hope it helps.

Simone