Update caches for federated shares in cron

[rullzer]

Currently we check the status of federated shares often during a request to list the files. Especially with broken federated shares this can lead to huge slowdowns. #7485 tries to fix this a bit.

However, it will still result in a request every once in a while to be slow.

A better approach would be to have a cronjob running for all (incomming) federated shares to check the status.

As discussed: @MorrisJobke @schiessle

[enoch85]

Just noticed this, which is related to this: #7513

[MorrisJobke]

Most likely nothing for 14 -> moving to 15.