Decision: Merge #2549 Over pr_review.py's Unrecognized Overview Blocks (#2480) #2556

Question

May #2549 merge to develop although scripts/pr_review.py reports unrecognized Copilot overview blocks on two of its earlier rounds?

State

• Hand a Spent Edit Budget to the PR, and Narrow a Blocked Promotion to Develop #2549 implements Keep an Unattended Run Moving: Hand a Spent Edit Budget to the PR, and Narrow a Blocked Promotion to Develop #2534 for handoff #2535, track auto-2534. Its head is f2e01102.
• Every check is green (12/12) and the merge state is CLEAN.
• Copilot reviewed three rounds. The head round reads **0 open findings**, a shape the script does read, and lists the two findings of the round before as resolved since last review.
• Copilot raised three findings as review threads, and all three are resolved:
  • check the promotion blocker before opening a promotion, fixed in 15139489
  • complete the promotion body's Fixes list, fixed in f2e01102
  • page through all decision issues, disproven, since the read already runs under picker step 1's full-page check
• pr_review.py wait ends UNRECOGNIZED_REVIEWER_OUTPUT because rounds 44284596 and 15139489 carry the collapsible <summary><strong>N open finding(s)</strong></summary> block, singular and plural. That is the shape #2480 already tracks. Read by hand, each block lists exactly the findings that are now resolved threads.
• The head round's verdict line says "A remaining behavior-description inconsistency warrants human review" without naming one, and it raised no finding.
• Merge Gate item 4 blocks on an unrecognized shape, and the script says merging over one is the maintainer's call, so the unattended worker parked rather than merging.
• Four local strict-review passes ran. The remaining local findings are style-class sentence-length remarks, listed in the pull request body.

Options

1. Merge Hand a Spent Edit Budget to the PR, and Narrow a Blocked Promotion to Develop #2549 to develop as it stands (recommended). The unread blocks were read by hand, every finding in them is a resolved thread, and the head round is in a read shape with no open finding. Effect: the next session on the auto-2534 lane merges it to develop and, under scope main, promotes it, closing Keep an Unattended Run Moving: Hand a Spent Edit Budget to the PR, and Narrow a Blocked Promotion to Develop #2534 when the promotion merges.
2. Hold Hand a Spent Edit Budget to the PR, and Narrow a Blocked Promotion to Develop #2549 until Recognize Copilot's Collapsible "N Open Findings" Overview Block in pr_review.py #2480 lands, then rerun pr_review.py wait. This keeps merges to digests the script fully reads. Effect: the lane waits on Recognize Copilot's Collapsible "N Open Findings" Overview Block in pr_review.py #2480, and every unattended worker meets this same block until it lands.
3. Ask for a human read of the "behavior-description inconsistency" first. Effect: the lane waits on that read, and any change it asks for goes through another review round.

This decision blocks pull request #2549. It does not block the open promotion.