fix(cargo-coverage-gate): eliminate unkillable format_delta mutant

[martin-kolinek]

🤖 Fixes the scheduled anvil-scheduled workflow failure: both scheduled-exhaustive (linux) and scheduled-exhaustive (windows) legs failed anvil-mutants-full with exit code 2 because of a single MISSED mutant:

crates/cargo-coverage-gate/src/render/mod.rs:89:23: replace < with <= in format_delta

Root cause

format_delta rounds a delta to one decimal place and then chooses the sign text via:

if rounded > 0.0 {
    format!("+{rounded:.1}pp")
} else if rounded < 0.0 {
    format!("{rounded:.1}pp")
} else {
    "0.0pp".to_owned()
}

By the time this code runs, the sub-precision guard earlier in the function has already ruled out |delta| < DELTA_DISPLAY_PRECISION - DELTA_BOUNDARY_TOLERANCE (~0.0999999999999). Rounding any delta whose magnitude already clears that bound to one decimal place can never produce exactly 0.0 (verified numerically at and around the boundary). That means the rounded < 0.0 branch is unreachable in practice, so mutating it to rounded <= 0.0 is an equivalent mutant — no test can ever observe the difference, which is exactly why cargo-mutants reported it as MISSED on every scheduled run.

Fix

Restructured the branch so the unreachable comparison no longer exists:

let rounded = (delta * 10.0).round() / 10.0;
if rounded == 0.0 {
    return "0.0pp".to_owned();
}
if rounded.is_sign_positive() {
    format!("+{rounded:.1}pp")
} else {
    format!("{rounded:.1}pp")
}

Checking rounded == 0.0 first (which is caught by the existing format_delta_renders_exact_match_as_zero test either way it's mutated) and then dispatching on is_sign_positive() removes the redundant </<= comparison entirely, since is_sign_positive isn't a target for cargo-mutants' operator-swap mutations and its true/false stub mutations are already caught by the existing positive/negative sign tests.

This file (crates/cargo-coverage-gate/src/render/mod.rs) is hand-written crate source, not a cargo-anvil generated file, so no template edit or cargo anvil regeneration was required for this change. I confirmed just anvil-fmt --fix, just anvil-readme --fix, and just anvil-spellcheck produce no diffs.

Verification

• cargo test -p cargo-coverage-gate — all 47 unit tests + 3 ignored + 1 doctest pass.
• cargo mutants -p cargo-coverage-gate --test-workspace false --file crates/cargo-coverage-gate/src/render/mod.rs --no-shuffle — 334 mutants tested: 294 caught, 40 unviable, 0 missed (previously 1 missed).
• just anvil-clippy — clean, no warnings.
• just anvil-spellcheck — clean.
• just anvil-fmt --fix / just anvil-readme --fix — no changes.

Not related to the currently open PRs #177/#178 — this is a standalone fix scoped to the scheduled mutation-testing failure.

[Copilot]

Copilot review overview

🟢 Approval recommended

The only finding is a non-blocking documentation nit, and validation reports zero missed mutants.

Review effort: Lite
Findings: None

What changed in this PR

This pull request refactors format_delta to eliminate an equivalent mutation while preserving coverage-delta rendering.

Changes:

• Handles rounded zero explicitly.
• Uses is_sign_positive() for nonzero values.
• Retains existing sign and boundary test coverage.

File	Summary
crates/​cargo-coverage-gate/​src/​render/​mod.rs	Refactors delta formatting; a minor documentation nit requests clarifying that only the zero case is unreachable.

---

💡 Configure MCP servers for context-aware, tailored reviews. Learn more in the docs.

[Codecov Comments Bot (codecov-commenter)]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 100.0%. Comparing base (9248a3b) to head (7be6624).
⚠️ Report is 2 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff            @@
##            main     #197      +/-   ##
=========================================
+ Coverage   98.1%   100.0%    +1.8%     
=========================================
  Files        309       16     -293     
  Lines      85048     1553   -83495     
=========================================
- Hits       83505     1553   -81952     
+ Misses      1543        0    -1543     

Flag	Coverage Δ
linux	100.0% <100.0%> (+1.8%)	⬆️
linux-arm	100.0% <100.0%> (+1.8%)	⬆️
scheduled	?
windows	100.0% <100.0%> (+1.5%)	⬆️

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

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

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