[server] Reduce DB healthcheck interval in starter Docker compose template

The museum container depends on the postgres container being up and the DB being
ready to accept connections. To enforce this dependency, we use the healthcheck
attribute.

See: https://docs.docker.com/compose/startup-order/

The value of the healthcheck interval was set to 1s since the default (30s)
caused each `docker compose up` to require at least 30 seconds on each startup,
which was prohibitive. The downside is that the healthchecks continue to run
beyond the startup phase too, and for small VMs, this caused a lot of
unnecessary CPU usage.

Thankfully, now Docker has a new option for a different healthcheck during the
start phase:

> start interval is the time between health checks during the start period. This
  option requires Docker Engine version 25.0 or later.

They were added in Docker compose 2.20.2, released an year ago (2023-07-19).

https://docs.docker.com/compose/release-notes/#2202

server/compose.yaml

@@ -49,9 +49,8 @@ services:
           "-U",
           "pguser"
         ]
-      interval: 1s
-      timeout: 5s
-      retries: 20
+      start_period: 40s
+      start_interval: 1s
     volumes:
       - postgres-data:/var/lib/postgresql/data
     networks: