The p2pool server daemon itself sort-of does this already, since it connects to multiple other nodes.  Having a node in your own network is the lowest latency solution for the vast majority of people, and even if a front end existed that would find the lowest latency public p2pool node for you, it would take about as much work to set up and maintain as the actual p2pool daemon takes.

I could see a use for this, though, if you couldn't guarantee the availability of your p2pool node even within your own network and didn't want to statically assign a list of backup nodes into your miner.  Still, it's one more point of failure, unnecessary complexity, and you'd still better have a static list in case your auto-connect front end fails.