feature request: Health check endpoint(s)

I'm deploying this in kubernetes and it would be helpful to have a health endpoint of some kind that I can use which will inform when the daemon is up and ready to go -- databases connected, etc.

stretch goal: This is probably more than you want to deal with, but best practices for kubernetes would indicate that it should also start returning a non-success code as soon as a SIGINT is detected but current requests should get handled before shutting down. the flow would be like this:

• Startup: "ready" check fails
• After DB and redis connections "ready" check succeeds
• On SIGINT "ready" check starts failing again but connections are still accepted.
• After some arbitrary timeout no new connections are accepted (this should be the interval that kubernetes checks the ready endpoint)
• Once all connections are handled or some additional arbitrary timeout expires the process exits

The goal with that flow is that kubernetes will stop serving traffic to a pod as soon as it it sees that it has stoped returning "ready". Thus with this flow existing connections will be handled while new connections will go to a new instance of the pod, making for a seemless changeover to a new pod version.

As I said, that's a "this would be neat to move towards enterprise class stuff" but even if you can just give me a simple "yep we're up and ready to go" check that would be super handy =] Right now I'm using the metrics port -- showing it as ready as soon as it responds there -- but that's not a guarantee that it's actually up.

No files or tests are named. Start by locating the daemon's startup, metrics-port, database/Redis connection, request-handling, and SIGINT shutdown entry points; then determine where a readiness endpoint can reflect connection state and graceful shutdown. Done means Kubernetes can distinguish startup, ready, draining, and stopped states without interrupting current requests.