Dear all,

for a nonlinear problem, I would like to work with a LU solver for the linearised system instead of the default Krylov solver.

I tried the following:

```
problem =NonlinearVariationalProblem(F,uc,bcs,j)
solver=NonlinearVariationalSolver(problem)
prm=solver.parameters
prm["newton_solver"]["error_on_nonconvergence"] = False
prm["newton_solver"]["linear_solver"]= "mumps"
```

But this didn’t work for me…