Add optional retry-on-timeout handling

[Poyraxx]

Summary

• add a retry_on_timeout flag to task and cron registration
• route worker-enforced timeouts through the normal retry flow when that flag is enabled
• document the new option and cover it with a task timeout retry test

Why

Issue #109 points out that worker timeouts currently fail immediately, which makes per-task retry behavior awkward when the built-in timeout is the thing that fires. This keeps the existing default behavior intact while adding an opt-in path for tasks that should retry after a timeout.

Root cause

Worker-enforced timeouts were converted straight into TimeoutError results, so they never entered the same retry path used by StreaqRetry.

Validation

• python -m compileall streaq tests

Closes #109.

[Graeme22]

Hi, thanks for the contribution and sorry for the delay in reviewing! I was a bit hesitant to add this initially but given 2 different people have asked for it I guess it's worthwhile. LGTM!