Occam's razor favors hypothesis (ii).
There is only one way for everybody to have consciousness; thus consciousness is just part of the definition of person. Figuratively, this is just one extra feature of the world, and is required in order to explain why other people seem to respond to situations similar to the way I do.
If some people have consciousness, I have to add an extra label to people, conscious vs. not, in order to describe the world. Figuratively, this is "number of people" additional features of the world.
It is these extra conscious/unconscious labels, to people who otherwise act equivalently (to me, and each other), that constitute the "plurality" that "posited without necessity".
The objection might be made that "but consciousness is itself complex". Suppose that that consciousness entails a plurality of size F (100,1000 whatever). Then under the model that different people are/aren't consciousness then each may or may not have this plurality of features, and thus has (6 billion)xF features. Whereas the "everybody is conscious" hypothesis just has F of them. You've still added a whole bunch of degrees of freedom by allowing otherwise indistinguishable individuals to have or not have consciousness.
However complex consciousness is, it is more complex to say some people do and some people don't have it, than to assume that everyone is equivalent in this regard, unless there is a good reason to infer otherwise.
We do infer that some people are unconscious, but we do so when there are observable behavioral differences between them and ourselves (who are necessarily conscious in order to be making the observations). Thus, representing that difference is "posited with necessity", i.e. for a good reason.