test(coat-338): Scenario B — merge master into feature (base-branch sync)

[anto-deepsource]

End-to-end test for coat PR #338 — verifies the preserved truthy case: when HEAD is a merge whose second parent IS an ancestor of master, coat still applies the merge-specific filter (so AI doesn't re-review already-on-master content).

Topology after the next push:

master  ─── A ─── C ────────                       (master 1+ commits ahead)
            │     ▲
            │     │ ancestor of master ✓
            │     │
feature ─── A ─── P1 ─── M     (HEAD after sync-merge push)
                          ▲
                          │ P1 = feature work

Steps now in progress:

1. ✅ This push at HEAD = P1 — triggers Analysis Update demo_code.py #1 (PR mode).
2. ⏳ Push merge M = "merge master into feature" — Analysis Remove debugger #2 (commit mode, base-branch sync).
3. ⏳ Pull logs; assert merge classification says "is an ancestor of base branch master (base-branch sync); applying merge-specific filter".

DO NOT MERGE — test PR for tracking coat #338.

[deepsource-development]

DeepSource Code Review

We reviewed changes in 9d1323c...93c3ff0 on this pull request. Below is the summary for the review, and you can see the individual issues we found as inline review comments.

See full review on DeepSource ↗

PR Report Card

Overall Grade   Focus Area: Reliability

Security   Reliability   Complexity   Hygiene

Code Review Summary

Analyzer	Status	Updated (UTC)	Details
Python		May 26, 2026 8:40a.m.	Review ↗
Secrets		May 26, 2026 8:40a.m.	Review ↗

[deepsource-development]

Local `sum` variable shadows builtin function, causing bugs

The local assignment to sum shadows the builtin Python function sum(), making it unusable in the current scope. This can cause confusing bugs or runtime errors if the actual function is needed for summation or iterable processing.
Rename the local variable from sum to a non-builtin name to restore access to the builtin sum() function and prevent name conflicts.