tornado: Fix logging of tornado activity level.

This logging was apparently broken when sorting imports; it's a fairly
unique thing in our codebase that this would be a problem.  Prevent
future regressions by adding this exception explicitly to the isort
configuration.

.isort.cfg

@@ -6,3 +6,5 @@ known_third_party = django, ujson, sqlalchemy
 known_first_party = zerver, zproject, version, confirmation, zilencer, analytics, frontend_tests, scripts, corporate
 sections = FUTURE, STDLIB, THIRDPARTY, FIRSTPARTY, LOCALFOLDER
 lines_after_imports = 1
+# See the comment related to ioloop_logging for why this is skipped.
+skip = zerver/management/commands/runtornado.py

zerver/management/commands/runtornado.py

@@ -9,22 +9,23 @@ from django.core.management.base import BaseCommand, \
 from tornado import ioloop
 from tornado.log import app_log
 
-from zerver.lib.debug import interactive_debug_listen
-from zerver.tornado.application import create_tornado_application, \
-    setup_tornado_rabbitmq
-from zerver.tornado.event_queue import add_client_gc_hook, \
-    missedmessage_hook, process_notification, setup_event_queue
 # We must call zerver.tornado.ioloop_logging.instrument_tornado_ioloop
 # before we import anything else from our project in order for our
 # Tornado load logging to work; otherwise we might accidentally import
 # zerver.lib.queue (which will instantiate the Tornado ioloop) before
 # this.
 from zerver.tornado.ioloop_logging import instrument_tornado_ioloop
-from zerver.tornado.socket import respond_send_message
 
 settings.RUNNING_INSIDE_TORNADO = True
 instrument_tornado_ioloop()
 
+from zerver.lib.debug import interactive_debug_listen
+from zerver.tornado.application import create_tornado_application, \
+    setup_tornado_rabbitmq
+from zerver.tornado.event_queue import add_client_gc_hook, \
+    missedmessage_hook, process_notification, setup_event_queue
+from zerver.tornado.socket import respond_send_message
+
 if settings.USING_RABBITMQ:
     from zerver.lib.queue import get_queue_client