A crash loop on FF-Cloud Small project is too slow to trigger safe mode

[hardillb]

Current Behavior

If you deploy a flow that crashes on start, it takes too long to restart and crash again to trigger the safe mode logic in the flowforge-nr-laucher

Expected Behavior

A crash loop should trigger the safe mode logic

Steps To Reproduce

• create small project on ff-cloud
• Deploy a flow that crashes on startup (see Inject node set msg property uses synchronous context calls nr-persistent-context#16)
• Watch the project logs

Environment

• FlowForge version: 1.2.0
• Node.js version: 16.x
• npm version: 8.x
• Platform/OS: k8s/linux
• Browser: Any