Add off-center plume temperature calculation

Currently, CFAST includes a plume centerline temperature calculation from 
McCaffrey's correlation. This should be enhanced in two ways:

1.  In the short term, we should remove the need to have the target location 
exactly above the fire.  If the location is reasonable close (0.25 m perhaps), 
the gas temperature should still be the calculated plume temperature, not 
instantly the surrounding layer temperature. This is a quick fix.

2.  In the longer term, we should implement the off-center correlation of 
McCaffrey.

Original issue reported on code.google.com by cfastdev@gmail.com on 11 Oct 2011 at 1:21

atwhaley / cfast

Add off-center plume temperature calculation #69