Daily account-usage refresh no longer stalls on large hosted instances

The daily account-usage refresh on Tuist Cloud was stalling on larger hosted instances: the fan-out job that updates each account’s cached monthly remote cache hit usage was blowing past PostgreSQL’s 65,535 bind-parameter limit once the backlog of stale accounts crossed ~7,281, which caused the bulk insert to fail repeatedly, drain no accounts, and grow every day until the job was exhausted. The worker now reads the full stable snapshot of accounts first and then inserts the per-account jobs in batches of 1,000, so hosted accounts get their monthly usage refreshed correctly again.