It's easy to forget awaiting `DownloadProgress` in the current `download` API

[matheus23]

iroh::blobs::Client::download returns a DownloadProgress which needs to be awaited, too, if you want to wait for the download to finish.

So node.blobs().download(hash, ...).await? is not enough to wait for a download to finish.
You also need node.blobs().download(hash, ...).await?.await? (or alternatively call .finish().await? on the DownloadProgress).

This should be documented in the .download function's docs.
I also suggested renaming .download to .start_download, but that seemed like a more controversial change.

[rklaehn]

The way it works in the new API is that all operations return a OpProgress struct which implements IntoFuture, so there is only 1 await in the case where you don't care for progress.