Update AWS server health warnings

[patudom]

Right now, if we make requests from, say, the educator dashboard, it asks for stage states that might not exist yet. This triggers a bunch of 404 errors from the server, which AWS flags as degraded health on the server. We should change the server settings so we don't get notified about things that aren't actual problems.

[Carifio24]

For the moment, I've updated the monitoring to ignore application 4xx responses. I'll keep investigating to see whether there's a more robust way to handle this.