The log_ can be safely truncated, but if admins use the traffic data provided to them in the admin, it will be reset. I've never run into that being a problem though and honestly wish they didn't exist since they are actually even useless where a reverse proxy is used. :)
I would suggest against truncating the quote tables.
The cron table is auto cleaned by the cron process. If it's bloated, you have problematic cron scheduler settings.
_tmp tables are MEMORY tables used for partial indexes. They should not be removed, but could be safely truncated assuming no index was running and no admins were signed into the back end.
_cl tables are a vital part of the Mview pattern which EE uses in it's indexers. Removing these would break all the triggers and ruin the incremental / async index process. I believe they can be truncated without causing issue.
In a migration, clearing log tables is one thing, clearing data tables is an entirely different story. Why mess with the data needlessly? It's pretty much begging for just one more thing to be at risk of going wrong during the migration. If it's being done as part of an upgrade (as opposed to server migration) then do it till your hearts content... In a staging environment, and test the tar out of and preflight everything before the moment of truth hits.