Maxima values scheme



The other day I contributed a linsolve_by_lu function.  It does
a goofy mtell on the condition number. We need something better.
I could introduce a global mat_condition, I suppose. But what I'd
like is a CL 'values' scheme for Maxima:

sol : linsolve_by_lu(m,b,floatfield);  <--ok
[sol, cnd] : linsolve_by_lu(m,b,floatfield);  <--also ok sets both sol and 
cnd

If anybody wants to work on this, that would be great.

Barton