cd2sbgnml: "active" state of the proteins should be converted to "active" state variable

[ibalaur]

The "active" state of the proteins should be converted to "active" state variable so it is a different specie in SBGN - for example, TYK2 input and TYK2 output should not be exactly the same molecules (attached)

[ibalaur]

[adrienrougny]

This could actually apply to all forms of EPNs (e.g. complexes and subunits too), as soon as they are active.
The inactive state of an EPN should also be transformed to a state variable, when the same EPN but with an active state exists in the map (to respect OVAV).