1) I deleted the Nek3D database reader since it has been superceded by the
   Nek5000 reader.

2) I applied a fix from Hank Childs for a bug in the contour filter, where
   it didn't specify a variable when it was getting an interval tree.  This
   resulted in it sometimes getting the wrong interval tree and producing
   incorrect results.  A specific example is when you create a pseudocolor
   plot of one variable and then apply an iso surface operator of another
   variable.



git-svn-id: http://visit.ilight.com/svn/visit/trunk/src@7925 18c085ea-50e0-402c-830e-de6fd14e8384