feature(github-workflow): Implement write-path chunking for IssueSyncer

Problem Statement

While PR #11114 successfully completed a one-shot data migration and updated 5 consumer recursive-readdir readers to support chunked paths for Issues, it did not update IssueSyncer.mjs#getIssuePath to write new active/archived issues to chunked paths. Currently, IssueSyncer continues to write to flat paths (issues/ and archive/), creating a write-path vs archive scope asymmetry.

Proposed Solution

Refactor IssueSyncer.mjs to utilize the chunked pathing algorithm for writes, ensuring new and updated issues correctly land in Math.floor(issue.number / 100) + 'xx' (or the padStart equivalent) directories. This aligns IssueSyncer with the newly implemented PullRequestSyncer and DiscussionSyncer chunked write-paths.

Additionally, this ticket should track the extraction of the pathing primitive (e.g. String(number).padStart(4, '0').slice(0, -2) + 'xx') into a shared utility (ai/services/github-workflow/shared/chunkPath.mjs) to prevent future drift across the 3 syncers.

Acceptance Criteria

• IssueSyncer.mjs uses chunked paths for all issue file creations and updates.
• Path generation logic is extracted to a shared utility and consumed by all 3 syncers (Issues, PRs, Discussions).
• Existing flat files (if any) are correctly moved to their chunked counterparts during sync.
• Integration and Unit tests verify the new paths.