fix: implement resource limits and optimize probe timing for petclinic pods

[asafchen-dig]

This PR addresses pod restart issues in the petclinic service by implementing proper resource management and probe configurations.

Changes made:

1. Added resource limits and requests to values.yaml:

   • requests: cpu=200m, memory=512Mi
   • limits: cpu=1000m, memory=1Gi

2. Updated probe configurations in deployment.yaml:

   • Added initialDelaySeconds: 30
   • Set periodSeconds: 10
   • Added timeoutSeconds: 5
   • Added failureThreshold: 3

These changes will help prevent pod restarts by:

• Ensuring proper resource allocation
• Implementing appropriate probe timing parameters
• Preventing unnecessary restarts due to aggressive health checks

Testing:

• Monitor pod status using 'kubectl get pods -n agentic'
• Verify pods are running stably without restarts
• Check pod restart counts remain at 0
• Verify application logs show no crash patterns
• Confirm health checks are passing consistently