Stop reading a rounded sixth decimal place as a decomposition bug

[blooop]

the_stages_account_for_no_more_than_the_total summed each stage's round6'd
seconds and compared that against a separately round6'd total with an exact
<=. Every number in the document is rounded on its own, so each part can round
up by half a microsecond while the whole rounds down by as much — the sum
then exceeds the total by arithmetic, with nothing charged twice, and the test
reported it as a decomposition bug.

It fired on #543 — Bump docker/login-action from 3 to 4,
a dependency bump that touches no Rust:

0.060256000000000004 against Document { total: 0.060256, stages: [
  StageReport { stage: "devpod-up", seconds: 0.030157, ... },
  StageReport { stage: "tools", seconds: 0.030099, ... }] }

What changed

The comparison now allows the precision the document is actually written to —
one half-microsecond per stage plus one for the total — rather than demanding
exactness the reporting never promised. Binary addition error is orders of
magnitude below that. A stage genuinely charged twice overshoots by a sleep, so
the check still catches what it exists for.

The comparison moved into stages_within_total so both sides of that boundary
get a deterministic test: the exact numbers from the failing run on one side, a
stage counted twice on the other. The flake is now a pinned regression instead
of a comment explaining a tolerance.

🤖 Generated with Claude Code

Summary by Sourcery

Make stage timing validation tolerate independent rounding while continuing to flag actual over-counting.

Bug Fixes:

• Prevent rounded stage timings from being incorrectly reported as exceeding the rounded document total.
• Retain detection of genuinely duplicated stage charges while allowing documented rounding precision.

Enhancements:

• Extract the stage-versus-total validation into a reusable predicate with precision-based slack.

Tests:

• Add regression coverage for the reported rounding-boundary failure and for detecting a stage counted twice.

[sourcery-ai]

Sorry @blooop, you've used your own review budget of 250,000 diff characters for the last 7 days.

You can request another review in 3 days and 9 hours by commenting @sourcery-ai review. Upgrade to get a review now.

[sourcery-ai]

Reviewer's Guide

The PR makes stage-total validation tolerant of the document’s independent six-decimal rounding while preserving detection of real stage overcounting, and adds deterministic tests for both boundaries using the exact regression values.

File-Level Changes

Change	Details	Files
Replace the exact stage-total comparison with a precision-derived tolerance that reflects independently rounded values.	Allow half a microsecond per reported stage plus one half-microsecond for the rounded total. Keep the check focused on detecting meaningful overcharging rather than floating-point or display-rounding artifacts. Move the comparison into a reusable helper used by the timing test.	rust/devlaunch-core/src/timing.rs
Add deterministic regression coverage for both rounding noise and genuine double-counting.	Pin the previously failing reported values as an accepted boundary case. Verify that adding a duplicated stage still fails by a margin far beyond the tolerance. Construct test documents directly to avoid timing-dependent sleeps.	rust/devlaunch-core/src/timing.rs

---

Tips and commands

Interacting with Sourcery

• Trigger a new review: Comment @sourcery-ai review on the pull request.
• Continue discussions: Reply directly to Sourcery's review comments.
• Generate a GitHub issue from a review comment: Ask Sourcery to create an
  issue from a review comment by replying to it. You can also reply to a
  review comment with @sourcery-ai issue to create an issue from it.
• Generate a pull request title: Write @sourcery-ai anywhere in the pull
  request title to generate a title at any time. You can also comment
  @sourcery-ai title on the pull request to (re-)generate the title at any time.
• Generate a pull request summary: Write @sourcery-ai summary anywhere in
  the pull request body to generate a PR summary at any time exactly where you
  want it. You can also comment @sourcery-ai summary on the pull request to
  (re-)generate the summary at any time.
• Generate reviewer's guide: Comment @sourcery-ai guide on the pull
  request to (re-)generate the reviewer's guide at any time.
• Resolve all Sourcery comments: Comment @sourcery-ai resolve on the
  pull request to resolve all Sourcery comments. Useful if you've already
  addressed all the comments and don't want to see them anymore.
• Dismiss all Sourcery reviews: Comment @sourcery-ai dismiss on the pull
  request to dismiss all existing Sourcery reviews. Especially useful if you
  want to start fresh with a new review - don't forget to comment
  @sourcery-ai review to trigger a new review!

Customizing Your Experience

Access your dashboard to:

• Enable or disable review features such as the Sourcery-generated pull request
  summary, the reviewer's guide, and others.
• Change the review language.
• Add, remove or edit custom review instructions.
• Adjust other review settings.

Getting Help

• Contact our support team for questions or feedback.
• Visit our documentation for detailed guides and information.
• Keep in touch with the Sourcery team by following us on X/Twitter, LinkedIn or GitHub.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 94.90%. Comparing base (c2b4ca7) to head (bc7c889).
⚠️ Report is 11 commits behind head on main.

Additional details and impacted files

Flag	Coverage Δ
python	42.98% <ø> (ø)
rust	95.20% <100.00%> (-0.22%)	⬇️

Flags with carried forward coverage won't be shown. Click here to find out more.

Components	Coverage Δ
shipped code (rust)	95.20% <100.00%> (-0.22%)	⬇️
harness and tooling (python)	42.98% <ø> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.