Release v3.79.0

Added

• cratis chronicle observers retry-failed-partitions <OBSERVER_ID> retries every failed partition of an observer in one operation. The kernel recovers a bounded batch at a time; the command returns once recovery is started and reports how many partitions were failed. Requires Chronicle 19.80.0 or later.

Changed

• cratis chronicle jobs stop and jobs resume now report why the kernel refused the request (job not found, already concluded, not running, not persisted, nothing took it forward) in text and JSON output, and exit non-zero. They no longer claim success when the kernel refused. The Workbench job actions show a refusal as a failure too.
• Updated to Chronicle 19.83.0 client packages.