Chache stored_event_count (on Issue and Projet)

"possibly expensive" turned out to be "actually expensive". On 'emu', with 1.5M
events, the counts take 85 and 154 ms for Project and Issue respectively;
bottlenecking our digestion to ~3 events/s.

Note: this is single-issue, single-project (presumably, the cost would be lower
for more spread-out cases)

Note on indexes: Event already has indexes for both Project & Issue (though as
the first item in a multi-column index). Without checking further: that appears
to not "magically solve counting".

This commit also optimizes the .count() on the issue-detail event list (via
Paginator).

This commit also slightly changes the value passed as `stored_event_count` to
be used for `get_random_irrelevance` to be the post-evication value. That won't
matter much in practice, but is slightly more correct IMHO.

phonehome/tasks.py

@@ -103,8 +103,10 @@ def _make_message_body():
             "project_count": Project.objects.count(),
             "team_count": Team.objects.count(),
 
-            # event-counts [per some interval (e.g. 24 hours)] is a possible future enhancement. If that turns out to be
-            # expensive, one thing to consider is pulling _make_message_body() out of the immediate_atomic() block.
+            # event-counts [per some interval (e.g. 24 hours)] is a possible future enhancement. We've already seen that
+            # such counts are expensive though, but if _make_message_body() is pulled out of the immediate_atomic()
+            # block (which is OK, since there is no need for some kind of 'transactional consistency' to register this
+            # simple metadata fact), then it might be OK to add some more expensive queries here.
         },
     }