[6.x] Avoid loading the whole container listing in Asset::exists()

[edalzell]

Asset::exists() and Asset::metaExists() answered via container()->files(), which unserializes the whole cached recursive listing. Asset::find() calls both, so resolving one asset reference on the front end loaded the full listing. On a 197k-file S3 container that is an 80 MB cache entry and OOMs a 1 GB worker.

Both methods now use the in-memory listing only when this request has already loaded it (AssetContainerContents::isLoaded()), and otherwise do a single disk()->exists() call (one S3 HEAD). CP listings load the listing first, so they keep their zero-extra-IO behaviour.

I skipped the suggested meta-cache shortcut: move() never forgets the old path's meta key, so it would report a moved-away path as existing. A disk check is always right.

Some tests asserted the raw cached listing keys and relied on exists() loading the listing before the .meta file was written. The listing now loads afterwards, so those expectations include the .meta entries, which the listing already filters out on read.

Closes #15607