[AIRFLOW-3125] Monitor Task Instances creation rates (apache#3966)

Montor Task Instances creation rates by Operator type.
These stats can provide some visibility on how much workload Airflow is
getting. They can be used for resource allocation in the long run (i.e.
to determine when we should scale up workers) and debugging in scenarios
like the creation rate of certain type of Task Instances spikes.

Author: mingyexia

airflow/models.py

@@ -5281,6 +5281,9 @@ def verify_integrity(self, session=None):
                 continue
 
             if task.task_id not in task_ids:
+                Stats.incr(
+                    "task_instance_created-{}".format(task.__class__.__name__),
+                    1, 1)
                 ti = TaskInstance(task, self.execution_date)
                 session.add(ti)