fix: increase Neo4j healthcheck timeout

[Yvanss971]

Description

The Neo4j container can be incorrectly marked as unhealthy during startup because the current healthcheck has a 5-second timeout.

On my system, Neo4j successfully starts, but:

cypher-shell -u neo4j -p password "RETURN 1"

takes around 12.5 seconds to complete.

This causes the healthcheck to time out even though Neo4j is actually running correctly, which prevents dependent services from starting with make prod.

Changes

• Increased the healthcheck interval from 5s to 10s
• Increased the timeout from 5s to 20s
• Added a 30s start period
• Kept 10 retries
• Explicitly defined the healthcheck as CMD-SHELL

Testing

Tested locally with Docker Compose:

• docker compose -f docker-compose.prod.yml config → YAML OK
• Neo4j container → healthy
• PostgreSQL → healthy
• Redis → healthy
• API → healthy
• App → healthy

The Neo4j healthcheck now succeeds on the affected system.

[dextmorgn]

Looks good, thanks !