fix: retry rejected lifecycle termination requests

[silent-cipher]

Summary

• Retry lifecycle-check termination requests up to three times when the SP returns TerminateServiceError.
• Wait 10 seconds between attempts and stop promptly if the job if aborted.
• Keep retries opt-in so existing terminateServiceSync callers retain single-attempt behavior.

Why

Transient SP chain-node failures can reject termination before anything is submitted on-chain, leaving lifecycle-check data sets active. Retrying these specific failures reduces leaked throwaway data sets.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.

[beck-8]

Non-blocking suggestion:

• Comment and documentation (apps/backend/src/data-set-lifecycle/data-set-lifecycle.service.ts:58) are too absolute: SDK's TerminateServiceError also covers other HTTP errors, and it cannot be claimed that "it is definitely not submitted to the chain" solely based on this type.