Optimize get_monthly_active_user_count by splitting up queries to avo…

…id usage of .distinct

osf/metrics/reporters/institution_summary_monthly.py

@@ -87,17 +87,10 @@ def get_monthly_logged_in_user_count(self, institution, yearmonth):
     def get_monthly_active_user_count(self, institution, yearmonth):
         institution_users = institution.get_institution_users().filter(
             date_disabled__isnull=True
-        )
+        ).filter(
+            Q(logs__created__gte=yearmonth.target_month(), logs__created__lt=yearmonth.next_month()) |
+            Q(preprint_logs__created__gte=yearmonth.target_month(), preprint_logs__created__lt=yearmonth.next_month())
+        ).values_list('id', flat=True)  # Retrieve only user IDs to minimize data loaded
+
+        return len(set(institution_users))
 
-        active_users = institution_users.filter(
-            Q(
-                logs__created__gte=yearmonth.target_month(),
-                logs__created__lt=yearmonth.next_month()
-            ) |
-            Q(
-                preprint_logs__created__gte=yearmonth.target_month(),
-                preprint_logs__created__lt=yearmonth.next_month()
-            )
-        ).distinct()
-
-        return active_users.count()