Improve performance of follow recommendation scheduler (#16159)

Express follow_recommendations in terms of account_summaries rather than
accounts, integrate filters that are unconditionally used, and materialize
the resulting view.

This should result in the bulk of the computation being performed only once
instead of **once per recommendation language**.

db/views/follow_recommendations_v02.sql

@@ -0,0 +1,34 @@
+SELECT
+  account_id,
+  sum(rank) AS rank,
+  array_agg(reason) AS reason
+FROM (
+  SELECT
+    account_summaries.account_id AS account_id,
+    count(follows.id) / (1.0 + count(follows.id)) AS rank,
+    'most_followed' AS reason
+  FROM follows
+  INNER JOIN account_summaries ON account_summaries.account_id = follows.target_account_id
+  INNER JOIN users ON users.account_id = follows.account_id
+  LEFT OUTER JOIN follow_recommendation_suppressions ON follow_recommendation_suppressions.account_id = follows.target_account_id
+  WHERE users.current_sign_in_at >= (now() - interval '30 days')
+    AND account_summaries.sensitive = 'f'
+    AND follow_recommendation_suppressions.id IS NULL
+  GROUP BY account_summaries.account_id
+  HAVING count(follows.id) >= 5
+  UNION ALL
+  SELECT account_summaries.account_id AS account_id,
+         sum(reblogs_count + favourites_count) / (1.0 + sum(reblogs_count + favourites_count)) AS rank,
+         'most_interactions' AS reason
+  FROM status_stats
+  INNER JOIN statuses ON statuses.id = status_stats.status_id
+  INNER JOIN account_summaries ON account_summaries.account_id = statuses.account_id
+  LEFT OUTER JOIN follow_recommendation_suppressions ON follow_recommendation_suppressions.account_id = statuses.account_id
+  WHERE statuses.id >= ((date_part('epoch', now() - interval '30 days') * 1000)::bigint << 16)
+    AND account_summaries.sensitive = 'f'
+    AND follow_recommendation_suppressions.id IS NULL
+  GROUP BY account_summaries.account_id
+  HAVING sum(reblogs_count + favourites_count) >= 5
+) t0
+GROUP BY account_id
+ORDER BY rank DESC