perf(helpers): fast-path type assertion in AsOops before errors.As

[samber]

Summary

• errors.As is reflection-based and AsOops is called once per layer by every chain-traversal helper (recursive, getDeepestErrorAttribute, collectMaps). When the wrapped error is directly an OopsError - the common case in oops chains - a plain type assertion answers in ~1ns.
• errors.As remains as a fallback for chains interleaved with non-oops wrappers (fmt.Errorf %w, custom Unwrap types).

Benchmarks

benchstat vs main (darwin/arm64, Apple M3, -count=6 -cpu=1):

                │ before.bench.txt │          after.bench.txt           │
                │      sec/op      │   sec/op     vs base               │
ChainTraversal        1611.5n ± 1%   870.1n ± 0%  -46.01% (p=0.002 n=6)
ErrorDeepChain         46.47n ± 5%   46.62n ± 6%        ~ (p=0.699 n=6)
ReflectionPaths        514.3n ± 2%   533.9n ± 1%   +3.81% (p=0.002 n=6)
geomean                337.7n        278.7n       -17.46%

                │ before.bench.txt │            after.bench.txt            │
                │       B/op       │     B/op      vs base                 │
ChainTraversal      3.922Ki ± 0%     1.422Ki ± 0%  -63.75% (p=0.002 n=6)

                │ before.bench.txt │           after.bench.txt           │
                │    allocs/op     │ allocs/op   vs base                 │
ChainTraversal       14.000 ± 0%     6.000 ± 0%  -57.14% (p=0.002 n=6)

Note: BenchmarkReflectionPaths shows a stable ~4% time delta with unchanged allocations even though it does not execute AsOops in its timed loop - consistent with a binary layout artifact rather than a logical regression.

Test plan

Verified with go build ./..., go test ./..., and go test -bench=BenchmarkChainTraversal -bench=BenchmarkErrorDeepChain -bench=BenchmarkReflectionPaths -benchmem -count=6 -cpu=1 ./benchmarks run before/after the change and compared with benchstat.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 90.18%. Comparing base (470a3a2) to head (474bac4).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #134      +/-   ##
==========================================
+ Coverage   90.17%   90.18%   +0.01%     
==========================================
  Files          15       15              
  Lines        1231     1233       +2     
==========================================
+ Hits         1110     1112       +2     
  Misses         96       96              
  Partials       25       25              

Flag	Coverage Δ
unittests	90.18% <100.00%> (+0.01%)	⬆️

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.

[Copilot]

Pull request overview

This PR optimizes AsOops by adding a fast-path type assertion before falling back to errors.As, reducing chain-traversal overhead when the error is directly an OopsError (the common case for oops-only wrapping chains).

Changes:

• Add a direct err.(OopsError) type assertion fast path in AsOops.
• Retain errors.As as a fallback to preserve behavior when chains include non-oops wrappers.
• Document the performance rationale inline to clarify why the extra branch exists.

---

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