GitLab

Nothing seems to guarantee, strictly speaking, that the split between
ApplyUpdates() vs PassiveApplyUpdates() maps to initial sync having been
done.

In practice, that's almost certainly the case, we some flaky tests do
suggest the assertion gets violated. Circumstances that could contribute
to this failure include polling sync cycles and failing configuration
cycles, which would all lead to a non-configuration cycle downloading
the initial data for a sync datatype, and that being passed via
ApplyUpdates().

Since the DCHECK is not guaranteed to hold, let's just relax it.

TBR=treib@chromium.org

Bug: 1002549
Change-Id: I901886cb34b68266a024c259328be204a6f88fa5
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/1807294Reviewed-by: Mikel Astiz <mastiz@chromium.org>
Commit-Queue: Mikel Astiz <mastiz@chromium.org>
Cr-Commit-Position: refs/heads/master@{#696903}