Active Channel Count metric not functioning correctly

[Jdgilby]

Theres a few issues with this one

1. If the stream segment is multiple features all features are counted separately even when part of the same channel. I don't see how to dissolve and clip in the background without a segment Id of some sort to make sure non-primary channels ( if there are multiple) are treated separately. (This makes me think this has to be dealt with in protocol digitizing best practices)
2. Sometimes the count is correct sometimes not. Cant figure this out. See below.
   
   
   
   You can see 9 is getting counted and is 1 segment crossing through the sample frame. 11 is only counting the primary channel. 10 the primary is split up but neither section is entirely within the SF. No non-primary channels are getting counted.
   
   Heres the JSON, if this is JSON related please assign to me and advise changes.

[KellyMWhitehead]

1. I think some of what is going on is that this calculation has a proportional component in it to avoid double counting features, (i.e. if the length clipped to the sample frame is greatest in that frame, then it gets the count). This is clearly not what we want for this metric, so we will need to set up property or create a separate count calculation for this.
2. We might get by with having a 'dissolve by' attribute for this metric. If we dissolve by primary and non-primary features, then in theory the primary features should create one geometry (as long as they are connected through proper digitization.) and non-primary features should separate into several geometries. Multithreaded systems might complicate this - we will need to test with several different scenarios.

This will not make the 1.0 release, but we should jump on this soon.