Fix Docker networking race condition on first startup

- Add health check to Traefik proxy with ping endpoint
- Make backend depend on proxy being healthy before starting
- Ensures Docker networks are fully initialized before backend binds ports
- Prevents connection timeout issues on first 'docker compose up'

This fixes a race condition where the backend container would start
and bind to port 8000 before Docker's network bridges were fully
configured, causing connection timeouts from the host machine.

The fix ensures the backend waits for the proxy (Traefik) to be
healthy before starting, which guarantees all network routes and
port forwarding rules are properly established.

Author: bigjohncodes

compose.override.yml

@@ -12,6 +12,12 @@ services:
     ports:
       - "80:80"
       - "8090:8080"
+    healthcheck:
+      test: ["CMD", "traefik", "healthcheck", "--ping"]
+      interval: 5s
+      timeout: 3s
+      retries: 3
+      start_period: 5s
     # Duplicate the command from compose.yml to add --api.insecure=true
     command:
       # Enable Docker in Traefik, so that it reads labels from Docker services
@@ -34,6 +40,8 @@ services:
       - --api
       # Enable the Dashboard and API in insecure mode for local development
       - --api.insecure=true
+      # Enable ping endpoint for health checks
+      - --ping=true
     labels:
       # Enable Traefik for this service, to make it available in the public network
       - traefik.enable=true
@@ -59,6 +67,9 @@ services:
     restart: "no"
     ports:
       - "8000:8000"
+    depends_on:
+      proxy:
+        condition: service_healthy
     build:
       context: .
       dockerfile: backend/Dockerfile