fix(core): score empty traces as incomplete

[JSap0914]

Summary

Fixes #19.

computeTraceCompleteness() treated an empty event array as fully complete by returning 100. That made an empty or failed trace look partially audited instead of marking trace completeness as missing.

This changes the empty-trace branch to return 0 and adds a regression test through computeRiskScore().

Verification

• RED before fix: bun test packages/core/src/scoring/index.test.ts -t "returns zero trace completeness" failed with Expected: 0, Received: 100.
• GREEN after fix: bun test packages/core/src/scoring/index.test.ts -> 16 pass.
• Manual API proof: bun -e "import { computeRiskScore } from './packages/core/src/scoring/index.ts'; const score = await computeRiskScore([], 'empty-run'); console.log(JSON.stringify({run_id: score.run_id, trace_completeness: score.components.trace_completeness, evidence_admission_score: score.evidence_admission_score.score}));" -> {"run_id":"empty-run","trace_completeness":0,"evidence_admission_score":50}.
• git diff --check passed.

Notes

I also ran broader local checks after building packages/schema:

• bun run --cwd packages/core test still has two unrelated failures in policy-audit and report tests.
• bun run --cwd packages/core typecheck still fails on pre-existing bun:test type resolution and unrelated implicit-any diagnostics.
• bun run lint reports existing repository-wide Biome formatting/lint drift.

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.