Instead of mixing two consecutive biquad orders with a linear xfade, the
code used the same weight for both signals to mix, due to a copy-paste
blunder. Brown-bag fix this now. Also change a little the threshold for
using 32-samples chunks to cater for steepness jumping from 0 to 4 (this
doesn't change anything for commonly used rates and is just for
robustness in never-happening cases anyway).
- Better formula for the lpf smoothness parameter. It is computed only
on init and rate change anyway...
- Only run as many Biquads as needed to save computing power, esp. when
using low steepness and parameters are not changing.
