FIX Increase health retries

[emteknetnz]

Issue #11

I've seen quite a lot of sporadic job failures due to unhealthy databases

Example

Waiting for all services to be ready
  /usr/bin/docker inspect --format="{{if .Config.Healthcheck}}{{print .State.Health.Status}}{{end}}" 18ea2fbb88191054825fb20d90a099b366ea97c619d67f0dbbf3cb45b9fb9c90
  unhealthy

https://github.com/creative-commoners/silverstripe-admin/runs/7194459514?check_suite_focus=true

It's not limited to mysql57. I've seen examples online with health-retries set to 10 so increasing for all databases to see if this helps resolve