NGA-Subduction Updates

• Reconcile any feedback from authors/developers
• Update python generated test results given concrete implementation changes (it may difficult to generate results for combinations of Alaska/Global and Cascadia basin scaling, discuss).
• Consider whether to condition subduction models generally on Ztor, Zhyp, or both. Moved to #178 (closed)
• Set Cascadia models to have basin effects; remove basin from Global implementations; create (basin) variants for Alaska
• Fix low Vs30 errors
• AG_20 sigma