feat(transform): identityOf — canonical Postgres-native object identity

[pyramation]

Summary

Adds the object-identity layer to @pgsql/transform — the canonical answer to "what object is this statement about?", derived purely from classifier facts and the parser's node taxonomy (CreateStmt, CreateTrigStmt, IndexStmt, ...), never from surface syntax like RangeVars.

• ObjectIdentity — { kind, schema, name, table? }; table scopes objects only unique per table (triggers, policies, indexes, constraints, seed data). This tuple is the key for dependency graphs, semantic diffing, and downstream naming.
• identityOf(facts: StatementFacts): ObjectIdentity | null — splits the classifier's table-qualified table.trigger names, recovers an index's table from its relation reference, reads extensions from facts.extension; null for grants/comments (they ride with the object they attach to).

Deliberately no path rendering here: change paths (e.g. pgpm module layouts) are derived projections owned by the packaging layer that consumes the identity — this package stays pgpm-agnostic. The pgpm naming spec (pathFor) lands separately in the constructive monorepo, keyed on this ObjectIdentity.

Reservations documented: function-overload signature disambiguation; constraint own-names (facts currently carry only the target table).

Link to Devin session: https://app.devin.ai/sessions/562099f8b3d94877b5ab57ec8bb90bec
Requested by: @pyramation

[devin-ai-integration]

🤖 Devin AI Engineer

I'll be helping with this pull request! Here's what you should know:

✅ I will automatically:

• Address comments on this PR. Add '(aside)' to your comment to have me ignore it.
• Look at CI failures and help fix them

Note: I can only respond to comments from users who have write access to this repository.

⚙️ Control Options:

• Disable automatic comment, CI, and merge conflict monitoring