Work around a beneficial race

When data are loaded, we notify waiting code before updating the cache.

This can confuse white-box tests that make assertions about the client
cache after a load, because the cache change may or may not have been
made yet.

src/ZEO/tests/drop_cache_rather_than_verify.txt

@@ -114,9 +114,6 @@ If we access the root object, it'll be loaded from the server:
     >>> conn.root()[1].x
     6
 
-    >>> len(db.storage._cache)
-    2
-
 Similarly, if we simply disconnect the client, and write data from
 another client:
 
@@ -138,8 +135,8 @@ another client:
 - Drops or clears it's client cache. (The end result is that the cache
   is working but empty.)
 
-    >>> len(db.storage._cache)
-    1
+    >>> len(db.storage._cache) <= 1
+    True
 
 (When a database is created, it checks to make sure the root object is
 in the database, which is why we get 1, rather than 0 objects in the cache.)