Release v3.1: Stale Recreations Fix & Pull Retry Logic

This release fixes critical issues with container recreations when updating images, suppresses background docker daemon inspect errors from leaking into cron logs, and adds exponential backoff retry logic for registry pulls to handle network glitches.

Bug Fixes:

• Stale Container Recreations: Solved issues where old container IDs became stale after earlier compose service restarts, leading to empty warnings ([WARNING] is a standalone container). Containers are now mapped and pre-cached securely during the scan phase before the update runs.
• Docker Inspect Error Suppression: Cleaned up the background recreation logs, ensuring failures to inspect deleted/stale container IDs do not leak raw error: no such object stderr dumps to the cron log.
• Deduplicated Service Restarts: Added tracking of already recreated Compose services to prevent redundant container updates when multiple containers share the same compose configuration.

Enhancements:

• Registry Pull Retry Logic: Added robust retry support for pulling Docker images. If a pull command fails (e.g., timeout or network registry glitch), NeXdocMan will attempt to pull the image up to 3 times with exponential backoff (5 seconds and 10 seconds delay), logging warning attempts to the primary log file /var/log/NeXdocMan/nexdocman.log.