ci(variant): gate PRs on variant + binding soundness (REQ-263)

[avrabe]

What (v0.28 slice 3 — final variant P1, DD-076)

No CI gate exercised the project's variants/bindings — the committed-broken full-desktop.yaml shipped because nothing ran it. Adds a step to the existing Traceability job (reuses the binary the validate step already built — no extra compile):

• rivet variant check --model --variant on every artifacts/variants/*.yaml — fails on UnknownFeature, unsolvable configs, etc.
• rivet validate --model --binding --strict-variants on artifacts/bindings.yaml — dangling artifact IDs (REQ-258), unknown constraint feature names (REQ-259), and unknown attribute keys (REQ-261) all fail here.

set -euo pipefail ⇒ any broken variant/binding fails the PR. Runs on every PR (Traceability isn't paths-gated), so breakage can't slip through a non-Rust change.

Verification

Ran the gate's exact commands locally with a fresh binary: all 3 variants + the binding validation exit 0 on the current (fixed) repo; a broken variant exits 1 (per the earlier review). ci.yml parses.

This completes v0.28 (REQ-260/261/262/263) — the variant-hardening P1 silent-drops. Next: cut v0.28.0.

Refs: REQ-263, REQ-258/259/261, DD-076

🤖 Generated with Claude Code

[github-actions]

📐 Rivet artifact delta

Change	Count
Added	0
Removed	0
Modified	1
Downstream impacted (depth ≤ 5)	0

Graph

graph LR
  REQ_263["REQ-263"]:::modified
  classDef added fill:#d4edda,stroke:#28a745,color:#155724
  classDef removed fill:#f8d7da,stroke:#dc3545,color:#721c24
  classDef modified fill:#fff3cd,stroke:#ffc107,color:#856404
  classDef overflow fill:#e2e3e5,stroke:#6c757d,color:#495057,stroke-dasharray: 3 3

Loading

Modified

ID	Changes
REQ-263

📎 Full HTML dashboard attached as workflow artifact rivet-delta-pr-712 — download from the workflow run.

_{Posted by rivet-delta workflow. The graph shows only changed artifacts; open the HTML dashboard (above) for full context.}

[github-actions]

⚠️ Performance Alert ⚠️

Possible performance regression was detected for benchmark 'Rivet Criterion Benchmarks'.
Benchmark result of this commit is worse than the previous benchmark result exceeding threshold 1.20.

Benchmark suite	Current: 85f6ef1	Previous: 961f3a3	Ratio
store_lookup/100	1914 ns/iter (± 12)	1540 ns/iter (± 10)	1.24
store_lookup/1000	24760 ns/iter (± 541)	18884 ns/iter (± 207)	1.31
store_lookup/10000	353594 ns/iter (± 4407)	288063 ns/iter (± 3947)	1.23
store_by_type/100	95 ns/iter (± 0)	77 ns/iter (± 0)	1.23
store_by_type/1000	96 ns/iter (± 1)	75 ns/iter (± 1)	1.28
store_by_type/10000	95 ns/iter (± 5)	77 ns/iter (± 1)	1.23
diff/1000	723954 ns/iter (± 4546)	572672 ns/iter (± 5133)	1.26
query/100	1294 ns/iter (± 4)	969 ns/iter (± 10)	1.34
query/1000	14797 ns/iter (± 84)	12084 ns/iter (± 149)	1.22

This comment was automatically generated by workflow using github-action-benchmark.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!