Agent retries job execution when Job Resource is missing instead of failing immediately

Current Behavior

• When a Job Resource is missing, the Agent treats the situation as a retriable failure.
• The Agent reports the error code ProcessLost, which is an indicator of retry.
• As a result:
  • The Agent repeatedly retries the execution.
  • The journal grows continuously due to repeated retries.
  • In long-running situations, this can lead to high memory consumption or Agent running out of memory.

Desired Behavior

• When a Job Resource is missing, the Agent should treat this as a non-retriable failure.
• The Agent should report the error code Failed  instead of ProcessLost.
• As a result:
  • No retries are performed by the Agent.
  • Related orders are immediately set to the Failed state.
  • The journal does not grow unintentionally.
  • The Agent remains stable and avoids unnecessary memory usage.