Introduce Ephemeral Storage Requests and Limits for Akash-node Helm Chart

[andy108369]

We've observed that akash-node can consume significant disk space over time, potentially leading to storage issues. To address this and prevent future occurrences, we propose introducing Kubernetes requests and limits for ephemeral storage in the Akash-node Helm chart. This would allow the node to be restarted automatically when it reaches the specified storage limits, improving resource management and stability.

Suggested configuration:

resources:
  requests:
    ephemeral-storage: "10Gi"  # Request 10 GiB of ephemeral storage
  limits:
    ephemeral-storage: "50Gi"  # Set a hard limit of 50 GiB of ephemeral storage

And these parameters should be configurable via the values.yaml.

https://kubernetes.io/docs/concepts/configuration/manage-resources-containers/#setting-requests-and-limits-for-local-ephemeral-storage

Rationale:

This configuration will ensure better control over disk usage by setting an initial request for storage and a hard limit to avoid excessive consumption. If the limit is reached, Kubernetes will automatically handle the pod lifecycle, preventing long-term storage issues. I.e. it'll simply bounce the pod and the node will quickly sync-up again.

Implementing this change will enhance overall reliability and resource management for Akash-node deployments.