use HEAD for sync

[alensiljak]

• I have searched for existing issues that may be the same as or related to mine.

I have not looked at the code but, seeing the synchronisation taking quite long for a small number of files, my assumption would be that the whole remote file is read for comparison. That has also been hinted in #8.

Therefore, I would suggest using the HEAD request instead of GET and reading the last-modified date/time from the response header.
This might significantly reduce the traffic and improve the performance of the sync operation.

As a workaround I will try syncing to a local folder and then performing the sync on the same collection, using rclone.