fix(check): accept a feature reference to a named multiplicity

[devin-ai-integration]

What and why

The feature-reference rule (passes/w8c_feature_reference.go, checkReferent) reported Must be a valid feature for a reference to a named multiplicity, because its test was isUsageKind(target.Kind) and SymbolMultiplicity is not a usage kind:

class E {
    multiplicity em [1..4];
    feature v : ScalarValues::Natural = em;   // was: error Must be a valid feature
}

The condition now reads !isUsageKind(target.Kind) && target.Kind != symbols.SymbolMultiplicity. SymbolKind.IsFeature() already classifies a multiplicity as a feature. A class or other type used as a referent still reports the error, and the accessibility check is unchanged.

This false positive appeared in the reproduction for Systems-Modeling/SysML-v2-Pilot-Implementation#802, which expects no diagnostics there. The docs now say that the maintainer confirmed #802 and fixed it upstream on ST6RI-975 (commit 1563e068, Type_multiplicity_SettingDelegate now derives multiplicity from ownedMember). The fix is not in a release yet. With that class compiled ahead of the pinned 2026-08 jar, the #802 reproduction and its subclass and struct variants validate clean.

Specification basis

KerML 1.1 §8.3.3.1.9: Multiplicity specializes Feature, so a FeatureReferenceExpression may name one. No row of docs/project/spec-compliance.md moves; the validation-constraints.md and omg-issues.md entries for #802 record the upstream status.

How it was verified

• New TestW8CFeatureReferenceToMultiplicity: an in-class multiplicity, a package-level multiplicity, and the full fix(repl): exit on Ctrl-C at an empty prompt, discard the continuation otherwise #802 reproduction, all with no errors. The existing class-referent negative control in TestW8CFeatureReferenceAccessibleAndValid still reports.
• Checked against the pinned pilot validate-kerml. The pilot accepts the package-level feature w = pm;. Its remaining diagnostics on the in-class case come from the fix(repl): exit on Ctrl-C at an empty prompt, discard the continuation otherwise #802 delegate bug, and they disappear with the ST6RI-975 class.
• go vet ./internal/check/..., make lint, go test ./internal/check/... ./internal/semantic/..., and the training and pilot corpus gates (OPENSYSML_REQUIRE_TRAINING_CORPUS=1 OPENSYSML_REQUIRE_PILOT_CORPORA=1) all pass with no rows moved. The changelog, doc-id and doc-link checks also pass.

Checklist

• make test and make lint pass locally (scoped go test above plus the corpus gates; the full suite runs in CI)
• Tests added or updated for the change
• Documentation extended where it already covers the surface (see CONTRIBUTING.md)
• Changelog entry added as changes/unreleased/<slug>.<section>.md, not as an edit to CHANGELOG.md
• baselines regenerated and make docs-counts run if a gate count moved (no gate count moved)
• No internal work-item labels (waves, slices, F4, K5) in the body, docs, or changelog

Link to Devin session: https://nasa-jpl-demo.devinenterprise.com/sessions/9c1dd2015da0409389e714253be0fb0b
Open in Devin Desktop: https://nasa-jpl-demo.devinenterprise.com/desktop/session/9c1dd2015da0409389e714253be0fb0b?variant=devin
Requested by: @HuiJun

[devin-ai-integration]

I'll fix CI failures and address comments from users with write access. I'll skip comments containing "(aside)".

• Disable automatic comment, CI, and merge conflict monitoring

[devin-ai-integration]

✅ Devin Review: No Issues Found

Devin Review analyzed this PR and found no bugs or issues to report.