Use phase pick times from Comcat when available

[kallstadt-usgs]

Is your feature request related to a problem? Please describe.
No.

Describe the solution you'd like
Phase pick information is available in Comcat in the quakeml.xml file, and these are often refined by expert human analysts. I suggest using the first arrival pick times when available within gmprocess for each station, when the pick time is available, rather than computing the phase arrival times using a simple earth model as is currently done.

Describe alternatives you've considered
None.

Additional context
None.

[mhearne-usgs]

Two things to check before we embark on a project to do this:

1. look at how often strong motion channels are used to pick phase arrivals. My experience tells me it's not often.
2. compare histograms of strong ground motion station distances with distances from stations used to locate/size earthquakes. For large earthquakes, the stations used (in this one example) range from 1.17 degrees (130 km) to 98 degrees (10,000 km). There will likely be some overlap, but I'm skeptical about how much.

https://earthquake.usgs.gov/earthquakes/eventpage/us6000iv6c/origin/phase