Clarify documentation for MaximalBlocks

[fingolfin]

The documentation for MaximalBlocks says that it

... returns a block system that is maximal (i.e., blocks are maximal with respect to inclusion) for the action of G on Omega. If seed is given, a block system is computed in which seed is a subset of one block.

This is confusing: first off, it could be read as saying that always the block system consisting of only MovedPoints(G) is returned (its unique block is maximal with respect to inclusion, after all). That's not what is meant.

Next, one might think that just the remark "maximal but not everything" is meant. Then for a primitive group, trivial size 1 blocks should be returned. But that's not the case either.

What seems to actually happen is that the code iterate to find bigger and bigger blocks until the action becomes primitive. That seems to fit with observed behavior.

While this is not the definition I would use (I find it surprising), it's what existing code now relies on, so I don't think we should change it. However, we definitely should clarify this in the documentation.