an entry page for the mm module

not only check for the number of bond partners, but also whether
there is an atom called SG in the bonded atoms.

of threads in the cpu platform

when it could be renamed in the meantime...

was just a copy paste artefact

can also read some CHARMM stuff

building disulfid bond

the settings object.

In many cases the user might only want to minimize a structure,
an integrator is therefore useless. To make OpenMM happy, a
dummy integrator (Verlet integrator with timestep 1fs) is
attached to the settings object and simulation in these cases.

In the topology creator, the input entity has been renamed to the
naming defined in the forcefield. After setting up the topology,
the naming does not matter anymore, as long as there are no
manipulations on the atoms. With the additional flag, the topology
creator can rename the entity back to pdb standard, as soon as
the whole parametrization magic took place.

Changed std::vec<geom::Vec3> into geom::Vec3List.

all terms separately

(Steepest descent and LBFGS)

Now it's possible to get the info back, whether steepest descent has
converged to the desired tolerance. Unfortunately this is not possible
for LBFGS. This is due to the implementation in OpenMM...

to make sure, that the angles get written properly into dcd files.

not to be reloaded (due to the previously applied copy to the entity)

 right new they are: SOL HOH TIP3

hydrogen constraints

check for first char equals H

This reverts commit bf46ad1a.

Conflicts:
	modules/conop/src/compound.hh
	modules/conop/src/compound_lib.cc

allows simple things like building disulfid bonds, heme bonds, ...,
and some renaming functionality