fix issue where signal instances were being related to the wrong port

[leviathan747]

The current code makes the assumption that there will only be one instance of SPR_[PO|PS|RO|RS] for a given instance of C_EP. This assumption fails when more than one component implement a given interface.

[coveralls]

Coverage decreased (-0.03%) to 81.505% when pulling f5ba983 on leviathan747:signals into 193bdbe on xtuml:master.

[john-tornblom]

I took a quick look at the metamodel and you appear to be correct.
Normally I would expect a test case for this, but defining stimuli in this region of the metamodel is cumbersome, and will grow out of control without a good strategy. See #12 (comment) for additional thoughts.