Support MPFR floats in numkondo

Remove '-D' option (error tolerance) in numkondo

NOTICE

@@ -1,2 +1,15 @@
 meankondo
 Copyright 2015 Ian Jauslin
+
+The numerical values can be represented as multi-precision floats using
+the GNU MPFR library, which is licensed under the GNU Lesser General
+Public License (LGPL) version 3 (see LGPL3 for a copy of the license).
+See
+  http://www.mpfr.org/
+for details.
+
+The GNU MPFR library is based on the GNU GMP library, which is licensed
+under the GNU Lesser General Public License (LGPL) version 3 (see LGPL3
+for a copy of the license).  See
+  http://www.gmplib.org/
+for details.