Rebase Vite/Vitest tooling onto our stack (keep MAX_BUFFER_SIZE fix)

[OBrezhniev]

Rebase tooling migration onto our stack (Vite + Vitest)

Applies feat/esm-migration tooling on top of our feature/sharedArrayBuffers
work. Target: feat/esm-migration.

• Rollup → Vite (node build/main.cjs + browser ESM), Mocha → Vitest with
  node + browser (Playwright) projects, flat eslint.config.js, CI.
• Keeps our src/binfileutils.js: the MAX_BUFFER_SIZE = 1 GiB constant.
  feat/esm-migration reintroduced the Buffer.constants.MAX_LENGTH probe,
  which is always undefined (those constants live on the buffer module, not
  the Buffer class) — the bug we fixed. Please take this version.
• Keeps our git+https pins (fastfile/ffjavascript) + audit overrides.

Validation: lint clean, node + browser round-trip tests pass.

[OBrezhniev]

ESM-rebase set — the tooling migration re-applied on top of our optimization stack (rather than replaying the branches), so the MSM/CIOS/memory work is preserved. Review together:

• Rebase ESM + Vite/Vitest onto CIOS/MSM stack wasmcurves#76 (full CJS→ESM; keeps our CIOS/GLV hot files)
• Rebase Vite/Vitest tooling onto MSM stack (build/test only, no threadman rewrite) ffjavascript#186 (Vite/Vitest only — WorkerSlot/batch-MSM threadman kept, no workerpool)
• Rebase Vite/Vitest tooling onto direct-io stack fastfile#137 · Rebase Vite/Vitest tooling onto our stack (keep MAX_BUFFER_SIZE fix) #89 · Rebase Vite/Vitest tooling onto deps-update stack r1csfile#106
• Rebase Vite/Vitest tooling onto prover stack (self-contained IIFE, external bfj) snarkjs#629 (self-contained IIFE + external-lazy bfj kept)

Deviations from feat/esm-migration are called out in each PR body (e.g. binfileutils MAX_BUFFER_SIZE, snarkjs IIFE/bfj, mnt6753 excluded, smart_contracts test deferred). All branches keep git+https commit-ref pins; re-pin if a branch moves.