perf: upgrade AG Grid v33 → v34

[paddymul]

Summary

• Version bump only — v34 has zero breaking changes
• ESM bundle: 1,722 KB → 1,773 KB (+3% code growth)

Test plan

• tsc -b — 0 errors
• pnpm test — 67/67 pass
• vite build — success
• CI checks pass

🤖 Generated with Claude Code

[github-actions]

📦 TestPyPI package published

pip install --index-strategy unsafe-best-match --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ buckaroo==0.12.9.dev22336319086

or with uv:

uv pip install --index-strategy unsafe-best-match --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ buckaroo==0.12.9.dev22336319086

MCP server for Claude Code

claude mcp add buckaroo-table -- uvx --from "buckaroo[mcp]==0.12.9.dev22336319086" --index-strategy unsafe-best-match --index-url https://test.pypi.org/simple/ --extra-index-url https://pypi.org/simple/ buckaroo-table

[paddymul]

Closing — v34 not needed

v33 gives us the best bundle size (-31% wheel, -61% widget.js) and v34 adds ~3% back with no features relevant to buckaroo:

• Cell editor validation, batch/bulk editing — we're read-only
• Tree data drag & drop, date grouping — not used
• AI toolkit / MCP server — not applicable
• React 19.2 support — not needed yet

No performance improvements for infinite scroll, pinned rows, or cell rendering.

Policy: don't bump AG Grid until we need a specific feature. Can always revisit if something in v34+ becomes relevant.