viz: reduce number of loops over overlays

Both SkiaOutputDeviceBufferQueue and OutputPresenter
loop through the list of overlays, which wastes CPU cycles.
We are basically doing O(n) x 2, which is inefficient.

This patch fixes that by passing overlay candidates and
their accesses one by one.

Bug: 1295527
Change-Id: Id6140467be6cffa5456f9da081b0fa53dc10a389
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/3689501


Reviewed-by: Peter McNeeley <petermcneeley@chromium.org>
Commit-Queue: Maksim Sisov <msisov@igalia.com>
Cr-Commit-Position: refs/heads/main@{#1012858}