chore(models): sync intellect scores from Artificial Analysis

[copse-release-bot]

New Artificial Analysis v4.3 measurements were disconnected from Copse’s canonical v4.1 intellect scale. Link them with a frozen affine map fitted to 611 models measured on both versions, so new models appear without changing existing canonical measurements.

Translated scores retain estimate and extrapolation provenance. The top-band representative and affected advisor/frontier fixtures now follow the updated data. The fit has R² 0.9933 and RMSE 1.31 on shared anchors; those are fit diagnostics, not guarantees for extrapolated models.

Validation: canonical-preservation and provenance regression tests; focused advisor/model/frontier tests; translated-score picker visual eval; five advisor-pair Electron cases with reviewed screenshots. Full pnpm run check passed: 8,745 tests passed, 7 skipped, none failed. Fresh CI is pending on the updated head.

[github-actions]

🖥️ PR preview

• Marketing site
• Browser demo
  Interactive, with a mocked backend — jump to a scenario:
• Builds a cupcake site
• Agent-proposed edits open the Changes panel
• Markdown list indentation
• Responsive composer footer
• Subagent display visual reference
• Subagent stays ungrouped beside reading tools
• Semantic search subagent markdown
• Light-theme approval with a bright accent
• Light-theme syntax, fills, and selection
• Grouped outside-sandbox command approval
• Remote desktop discovered-port list with one selected
• Settings scroll + sticky footer geometry
• Per-model reasoning / temperature / top-p
• Thread switch hydration notice
• Thread switch hydration failure
• Chat layout styling
  _{Static demo of the renderer against a mock API — the browser pane, terminal, and all main-process IPC are stubbed, so it never proves main-process behaviour. Rebuilt whenever a push changes the demo bundle; these links go live once the queued deploy lands, shortly after this comment. Removed when the PR closes.}