fix(pgpm): ignore PG17+ list offsets in the AST round-trip check

[pyramation]

Summary

pgpm package warns ⚠️ AST round-trip diff detected! on nearly every module. Almost all of it is noise: PG17+ added positional fields that cleanTree never stripped — A_ArrayExpr.list_start/list_end and A_Expr.rexpr_list_start/rexpr_list_end. Any ARRAY[...] or x IN (...) shifts those byte offsets when the SQL is reformatted, so the reparsed tree never matches. Scanning constructive-db: 3775 + 78 offset-only differences across 30 modules, hiding exactly one real deparser bug (a dropped EXCLUDE constraint name — fixed in constructive-io/pgsql-parser#331).

The diff check now uses a separate cleanTreeForDiff that also drops those four keys. cleanTree is deliberately left alone because hashSqlFile uses it for pgpm_migrate ledger hashes — widening it there would rewrite the hash of every already-deployed change containing an array.

Link to Devin session: https://app.devin.ai/sessions/5cfcc6690b574cdd96de080ee3b94b4e
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