fix(actions): run TransferLogs on UpdateLog{Rows:[], NoMore:true}

[silverwind]

UpdateLog short-circuits on len(Rows)==0 before honoring NoMore, so a final empty UpdateLog{NoMore:true} never runs TransferLogs. The task's dbfs_data rows are then never moved to log storage and never deleted.

The bug has been latent since the original Actions implementation, act_runner versions after runner#819 trip it deterministically.

Fix: let NoMore=true with no new rows fall through to TransferLogs. Bail when the runner has outrun the server (Index > ack) even with NoMore, since archiving a log with a gap is worse than retrying. Always call WriteLogs so offset==0 bootstraps an empty DBFS file in the no-output case (otherwise TransferLogs would fail at dbfs.Open).

Fixes: #37623
Ref: runner#952
Ref: runner#950

---

This PR was written with the help of Claude Opus 4.7

[Copilot]

Pull request overview

Fixes a long-standing Actions runner log finalization edge case where a final UpdateLog{Rows: [], NoMore: true} request could be short-circuited, preventing TransferLogs() from running and leaving dbfs_data rows orphaned (unbounded DB growth).

Changes:

• Extracted the log-update decision logic into planLogUpdate() to correctly handle empty finalization requests and “runner ahead of server” scenarios.
• Updated UpdateLog to write only the unacked rows and to allow NoMore=true with no new rows to still finalize/transfer logs.
• Added unit tests covering planLogUpdate() behavior, including the regression scenario.

Reviewed changes

Copilot reviewed 2 out of 2 changed files in this pull request and generated 1 comment.

File	Description
routers/api/actions/runner/runner.go	Refactors UpdateLog control flow via planLogUpdate() and ensures NoMore requests can trigger finalization even when Rows is empty.
routers/api/actions/runner/runner_test.go	Adds unit tests for planLogUpdate() including regression coverage for empty finalization behavior.

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.

[Copilot]

Pull request overview

Copilot reviewed 2 out of 2 changed files in this pull request and generated no new comments.

[bircni]

@silverwind you wanna backport this?

[silverwind]

Could backport but not strictly needed because the runner fix is merged and serves as a acceptable workaround. If we backport, we could remove the runner workaround with 1.27, so imho still valuable to not have that workaround laying around for 3+ months.