Release v3.2: Dynamic Retry Logging & Log Cleanup

Release v3.2: Dynamic Retry Logging & Log Cleanup

This release resolves false-positive error logging during transient container registry pull retries. Failed pull attempts that are successfully retried are now logged with a warning severity instead of polluting logs with critical exit code dumps.

Enhancements:

• Dynamic Command Retry Logging: Configured run_command to support custom error severities using the RUN_COMMAND_ERR_LEVEL environment variable.
• Cleaned Retries: Docker pull failures on transient attempts (e.g., rate limits or registry timeouts) are now flagged as [WARNING] in log files rather than [ERROR], preserving the high-priority [ERROR] logging only if the final attempt fails.