Technically this would work, but I don't see much point.  At the Br absorption peak you'd expect an f'' of at least 3.8e- for Br, versus 0.2e- for the sulfurs, which means that Br should be by far the largest peak in the map unless it's just not there at all.  Refining anomalous scattering the sulfurs won't change this.  (What the LLG map is useful for is the reverse - refine the anomalous scattering for Br, and the LLG map will more clearly show the sulfurs.)

Yes, the anomalous 0.92Å-minus-1.0Å map would be expected to show a positive peak around any Br atoms (it works for SeMet MAD data, at least).  But again, I think if it is bound at all the conventional anomalous difference map for the 0.92Å data should be clear enough if the ligand is bound at high enough occupancy to be modeled accurately.

-Nat