Reduce accepted Setline code-health debt

[sarthakagrawal927]

Follow-up for the initial Fleet code-health baseline in #27. The required gate must not be weakened while this accepted debt is reduced:

• Coverage: 60 web/Worker tests cover 3,544/10,739 production lines/statements (33.00%), 632/839 branches (75.32%), and 109/163 functions (66.87%). Ten native unit tests and four UI tests cover 3,245/4,962 production Swift lines (65.3970%).
• Complexity: 24 threshold violations across 470 TypeScript, JavaScript, and Swift functions / 6,176 function NLOC; maximum CCN 52, function length 528, and 14 parameters.
• Duplication: 6 exact clone groups / 56 duplicated lines / 0.3225% across 39 web/native production files.
• Dependencies: 16 accepted high advisory IDs (19 path findings), all pinned in the gate. Most are inherited through Next/React Server Components/Vinext/ESLint/C8; production-package upgrades require explicit owner approval. New critical/high IDs fail CI.
• Native unused analysis: Xcode compilation catches local unused values, while Knip is zero for the web/Worker surface; a pinned Swift declaration-level analyzer remains a gap.

Review by 2026-09-12. Cleanup receipt: run_669cade644da2a42ad7eace42d993dd8.

[sarthakagrawal927]

The accepted debt baselines are now enforced by CI with no new regression allowed: web/Worker coverage 33.00% lines/statements, 75.32% branches, 66.87% functions; native coverage 65.3970%; complexity 24 legacy violations (max CCN 52, length 528, params 14); duplication 6 groups / 56 lines / 0.3225%; 16 accepted high advisory IDs across 19 findings. Final hosted run: https://github.com/Significant-Hobbies/setline/actions/runs/31567132929.