tuning/debugging KDE output

[berl]

great new feature #27 gives smoothed outputs to spatial_detection_score and the results differ in some ways from the histogram. figure out the source of this difference and accommodate any real differences in an easy-to-understand visualization

[jsicherman]

My guess at the differences you see in #27, without any evidence, is that it's a plotting artifact from using fixed ranges