Always use the reduction of the maximum of the weighted residual as convergence criterion

This should make things a much more robust, partially because now the
linear and the non-linear solvers use the same convergence criterion.

Also, this patch includes some collateral indentation improvements.

examples/co2injection_flash_ecfv.cpp

@@ -55,7 +55,7 @@ SET_TYPE_PROP(
 #if HAVE_QUAD
 SET_TYPE_PROP(Co2InjectionFlashEcfvProblem, Scalar, quad);
 #else
-SET_SCALAR_PROP(Co2InjectionFlashEcfvProblem, NewtonRelativeTolerance, 1e-5);
+SET_SCALAR_PROP(Co2InjectionFlashEcfvProblem, NewtonTolerance, 1e-5);
 #endif
 } // namespace Properties
 } // namespace Opm