Inconsistency in the Writing/Reading of Bonds

[tonimueller]

The class WriteMcs skips the linearization of chains, if the connection is a valid bonds plus a multiple of the box size

LeMonADE/include/LeMonADE/core/MoleculesWrite.h

Line 151 in 7523a6b

else if ( (dummyPosition-molecules[n])*(dummyPosition-molecules[n]) >10 )

On the other hand, the two linear parts are written on the same line and in addition the bond is written to the bfm file in the function
void WriteAddBonds<IngredientsType>::writeStream(std::ostream& strm) . In this case, the bond is written two times to the file, although only one connection happened.

In the current reading process a second connection between the same monomers is not taken into account

LeMonADE/include/LeMonADE/core/ConnectedDecorator.h

Line 218 in 7523a6b

if (((uint32_t)b) == links[i]){

For consistency I would be good to only write the additional bond in !add_bonds if it is not taken into account by the linearization.

For this situation a test where two monomers are connected should be added (in the same box, apart by a periodic box).