It would be a huge advantage to have the most important cue points (in, out, crossfade) set automatically. As you have implemented the waveform feature already, it should be easy to use the "data" part of it.
Cue in/out would trim silence at the beginning and end of the track based on a configurable volume threshold (in/out each). The same can be done for the crossfade cue once the volume falls below a given level.
Auto cue detection should be an option which can be turned on in the station settings.