Color weirdness with model exports to FPP effect sequences (eseq)

[CaptainMurdoch]

There might be one more issue I need to look at FPP code again.  The step size needs to be rounded but the model size doesn't, but I think right now FPP reads model size as step size since the thought was that there is only one model but that logic is wrong.  We need to read step size as step size and use model size as the channel count to copy.  I will take a look at both and get it straight.

[Setarcos]

Thanks again Captain,

Let me know if I can help in testing.

[CaptainMurdoch]

OK, I just pushed simultaneous fixes to the FPP code and xLights code.  I needed to remove a line I added to xLights earlier which rounded the size of the model up to a multiple of 4.  As I stated after I thought about it, this part of my change didn't need to happen, but the original bugfix was still needed.  The negative effect of this new bug would be that FPP would corrupt 1-3 channels of data after the end of a model during playback of the effect.  I also modified FPP to honor the model size value, so FPP reads in the step size number of bytes, but only uses model size number of bytes when copying data to the channel output array.  This should be all fixed now.  FPP will work fine with xLights in the mean time, with the minor exception of those 1-3 extra channels of data being overwritten, so the pixel after your model may have bad data during playback until the effects are re-saved with the next xLights release.