Better kube liveness probe page

[Dragon1320]

So basically, it seems like the kube liveness probe for web is pinging the front page which makes a bunch of unnecessary requests to the api every time.

If that's not whats causing these extra requests then investigate what is.

Also while we're at it, add proper liveness routes, like /liveness or whatever to all apps which actually reflect their liveness (unlike api currently), and tweak the yaml liveness params, these routes can be used for uptime monitoring as well.

[Dragon1320]

For liveness checks, use the /healthz route and return 2xx or 3xx codes for success and 4xx or 5xx codes for errors. The route should perform any logic required to assess the overall health of the app and return the appropriate code. This should be as lightweight as possible as the route will be pinged frequently.

For the moment for api, check the database connection when pinging the health route as not doing so may not alert downtime using our current alert system (uptime robot).