...still buggy and very incomplete.

The new test does not pass yet but it might already work.

...as suggested by Bernd + Stefan

...also in this commit: Some minor cleanup

The fixed bug was the ignoring of the ldaU XML element if the version number
of the inp.xml file was not 0.29. For the NiO_ldauXML test this meant that
no linear mixing was used for the density matrix.

Please note that there are still slight differences between the results
obtained with this version and the last release version. I changed the
reference value for the total energy from -3191.9717 to -3191.9707.

This commit introduces general LDA+U parameters (for mixing) to the inp.xml file.

...hope, it works.

The k point density can now be defined by specifying denX, denY, denZ
parameters in the &kpts line.

Fixes to make pbe faster (again after refactoring of xcpot) Changes for reanabling old CPP_66 functionality

-removing of variable icorr, use of member-functions instead
-removal of icorrkeys module
-removal of hybridmix module
-cleanup at several places

It can now be specified according to the following example:

&atom element="Na" id=11.1 name="mySodium" /

This is the first step to enable the parallelization for parts of this code
as it is provided by Uliana.

Now fleur.F90 calls a subroutine that does most of the hybrid stuff.
In eigen.F90 only the non-local potential is added to the Hamiltonian

This is not yet tested.

Due to the elimination of variables some statements appeared redundantly.
I removed them.