Bug 1221536 - Docs: Emphasise starting the worker before ingest_push

If the worker is not running, any `apply_async()` calls are silently
thrown away, due to `ingest_push`'s use of `CELERY_ALWAYS_EAGER` and:
https://github.com/celery/celery/issues/2910

As such, running the worker after ingest_push doesn't help (since the
rabbitmq queues are empty) and so if people are interested in perf/log
data, then they must start the worker first instead.

docs/installation.rst

@@ -100,9 +100,10 @@ talos jobs for a particular push, try:
 
      (venv)vagrant@local:~/treeherder$ ./manage.py ingest_push --filter-job-group T mozilla-inbound 63f8a47cfdf
 
-Note that some types of data (e.g. performance) are not processed immediately, and you
-will thus need to start a celery worker to handle them. You don't need
-to enable the beat service for this though, so you can omit the "-B":
+Note that some types of data (e.g. performance, log error summaries) are not processed
+immediately, and you will thus need to start a celery worker *before* running `ingest_push`
+to handle them. You don't need to enable the beat service for this though, so you can
+omit the `-B`:
 
   .. code-block:: bash