Remove sideg list from sync return value

[sim642]

Previously sync transfer functions returned both the updated state (like every other transfer function), but also an assoc list of sidegs to perform. Since sync may just use sideg directly to make the same side effects to global unknowns, the returned list provides no additional value. It's there for historical reasons.

Therefore I see no value keeping it around. Its removal simplifies a lot of sync code, especially in BasePriv and Constraints.

[sim642]

One might have the performance concern that by doing sideg directly instead of collecting them causes some solver side effect functions to fire immediately, not at the end of the sync transfer. However, this doesn't actually happen because MCP2 for whatever reason collects sidegs anyway and does them together afterwards, including ones from different analyses:

analyzer/src/analyses/mCP.ml

Lines 837 to 868 in 06eb5ae

	let sync (ctx:(D.t, G.t, C.t) ctx) reason =
	let spawns = ref [] in
	let splits = ref [] in
	let sides = ref [] in
	let emits = ref [] in
	let f post_all (n,(module S:MCPSpec),d) =
	let ctx' : (S.D.t, S.G.t, S.C.t) ctx =
	{ local = obj d
	; node = ctx.node
	; prev_node = ctx.prev_node
	; control_context = ctx.control_context
	; context = (fun () -> ctx.context () |> assoc n |> obj)
	; edge = ctx.edge
	; ask = query ctx
	; emit = (fun e -> emits := e :: !emits)
	; presub = filter_presubs n ctx.local
	; postsub= filter_presubs n post_all
	; global = (fun v -> ctx.global v |> assoc n |> obj)
	; spawn = (fun l v a -> spawns := (v,(n,l,a)) :: !spawns)
	; split = (fun d es -> splits := (n,(repr d,es)) :: !splits)
	; sideg = (fun v g -> sides := (v, (n, repr g)) :: !sides)
	; assign = (fun ?name _ -> failwith "Cannot \"assign\" in sync context.")
	}
	in
	n, repr @@ S.sync ctx' reason
	in
	let d, q = map_deadcode f @@ spec_list ctx.local in
	do_sideg ctx !sides;
	do_spawns ctx !spawns;
	do_splits ctx d !splits;
	let d = do_emits ctx !emits d in
	if q then raise Deadcode else d

---

I noticed in the past (while implementing and tracing #183) that even though MCP2.do_sideg does this grouping, it didn't and still doesn't seem to properly work. I've identified a tricky interaction between ctx and query which is behind this phenomenon (some side effects bypassing the collection and going to the solver immediately. Since that's an orthogonal issue, I'll be opening a separate PR.