fix: canonicalize keywords and whitespaces

[Snehil-Shah]

Currently, many values are stored as spans of the input in the AST, which later produces a different digest for the same query written with different casing, spacing, or comments.

This PR stores the canonical spelling at parse time instead of deriving it from the input, with the parser writing its own separators rather than inheriting whatever sat between tokens. The result is stable ASTs and stable digests.

It also adds dialect-aware canonicalization for function names: MySQL resolves them case-insensitively (including backtick-quoted), while PostgreSQL folds only undelimited names, and folds them down.

[coderabbitai]

Important

• 🔍 Trigger review

This repository does not receive automatic reviews because it has fewer than 10 stars.

⚙️ Run configuration

Configuration used: Organization UI

Review profile: CHILL

Plan: Pro Plus

Run ID: 2e3be2aa-28a1-4926-9970-2e16246c20f5

---

_{Comment @coderabbitai help to get the list of available commands.}

[Snehil-Shah]

@renecannao I added a dedicated file as an AST-level regression suite for this. Let me know if you want it spread across existing files instead.