You are correct about the differences between cis/trans in alkenes vs. cyclic structures. However, to understand the use of the same descriptors, it helps to look at the similarities.
1. For both, you would have to break bonds in order to go from one isomer to the other. Even though there is some bond rotation in the cyclic structures, these rotations will never convert between cis and trans isomers.
2. In both cases, the substituents can either be on the same side (cis) or on opposite sides (trans) of a defined plane.
For the alkenes, this plane is a cross-section through the double bond that bisects the substituents on each carbon, for the cycloalkanes, the plane is defined by the atoms of the ring (even if they aren't truly planar as in a chair conformation, they can be drawn as if they were flat).
I've shown simple examples below where the plane in both cases is shown in light blue. I'm wishing I'd picked a different color considering the page background...let me know if it's not clear.