Snapshot mode: bundle SQLite snapshot for zero-infra usage

[dkijania]

Why

The README's roadmap already calls this out: snapshot mode currently requires the user to bring up Postgres via docker-compose, which is the single biggest onboarding cliff in the project. Bundling a SQLite snapshot inside the npm package would collapse snapshot mode into the same `npx`-and-go story as live mode.

Scope

• Build a SQLite version of the archive schema that the snapshot-mode tools can read.
• Add a CI job that periodically (weekly?) regenerates a fresh devnet snapshot from the existing Postgres dump and publishes it as a separate npm package (`@o1labs/mina-mcp-snapshot-devnet`) or a CDN/GCS download.
• Wire snapshot mode to:
  • Prefer a bundled SQLite if present.
  • Fall back to Postgres if the user explicitly points at one (existing behavior).
• Update the README so the snapshot-mode quick-start is one command, no Docker.
• Per-network variants (mainnet, mesa) follow the same pattern.

Open questions

• Snapshot file size — is a useful devnet slice within npm's package-size sanity limits, or does it have to be a separate download fetched on first run?
• Read-side compatibility — does `query_archive_sql` need rewriting for SQLite dialect differences, or do we keep the queries small enough that they're portable?

Acceptance

• `npx @o1labs/mina-mcp-server --mode snapshot --network devnet` works end-to-end with no Postgres, no Docker, no clone.
• Snapshot freshness is documented ("data is at most N days old") and visible at runtime via a tool like `describe_state`.

Related

• README `## Roadmap` section currently lists this as future work.