feat(lint): port the SQL lint engine

Add the sqlquack-lint crate: SQLClean's lint rules, ported from its Polyglot
engine onto SQLQuack's tokens and parse trees, so SQLClean and SQLBuild lint
without Polyglot. Snowflake and DuckDB lint natively; dialects SQLQuack does
not parse report lint as unavailable instead of failing.

sqlquack-syntax gains two public entry points the engine needs:
dialect_tokens, the token stream of a dialect, and parse_expressions, which
reads already guarded tokens as expressions. The rule walkers ported from
SQLClean keep its MIT notice in THIRD_PARTY_NOTICES.

Author: kvlonge

Cargo.lock

@@ -16,6 +16,7 @@ members = [
     "crates/sqlquack-syntax-postgres",
     "crates/sqlquack-format",
     "crates/sqlquack-format-check",
+    "crates/sqlquack-lint",
     "crates/sqlquack-syntax-bigquery",
     "crates/sqlquack-fingerprint",
     "crates/sqlquack-python",
@@ -51,6 +52,7 @@ polyglot-sql = { git = "https://github.com/sqlstack-io/polyglot.git", rev = "7d5
 proc-macro2 = "1"
 quote = "1"
 rayon = "1"
+rustc-hash = "2"
 serde = { version = "1", features = ["derive"] }
 serde_json = "1"
 # The Python module; abi3, so one build loads into every CPython from 3.12.
@@ -63,6 +65,7 @@ sqlquack-duckdb = { path = "crates/sqlquack-duckdb", default-features = false }
 sqlquack-evidence = { path = "crates/sqlquack-evidence" }
 sqlquack-format = { path = "crates/sqlquack-format" }
 sqlquack-fingerprint = { path = "crates/sqlquack-fingerprint" }
+sqlquack-lint = { path = "crates/sqlquack-lint" }
 sqlquack-policy = { path = "crates/sqlquack-policy" }
 sqlquack-rules = { path = "crates/sqlquack-rules" }
 sqlquack-semantics = { path = "crates/sqlquack-semantics" }

Cargo.toml

@@ -63,6 +63,8 @@ SQLQuack's DuckDB result-type rules, in crates/sqlquack-semantics/src/lineage/_h
 and duckdb_types.rs, are ported from polyglot's engine type rules, and the fixture
 crates/sqlquack-semantics/src/lineage/tests/fixtures/duckdb_engine_types.json is polyglot's
 tests/fixtures/engine_types/duckdb.json, the result types DuckDB bound for its synthetic cases.
+The lint Rules' scope walk and aggregate classifier, in crates/sqlquack-lint/src/tree, are ported
+from polyglot's scope.rs and traversal.rs.
 
 polyglot is distributed under the following licence:
 

THIRD_PARTY_NOTICES

@@ -0,0 +1,20 @@
+[package]
+name = "sqlquack-lint"
+version.workspace = true
+description = "SQL lint Rules over SQLQuack's tokens and syntax trees, ported from SQLClean."
+edition.workspace = true
+license.workspace = true
+repository.workspace = true
+homepage.workspace = true
+publish.workspace = true
+rust-version.workspace = true
+
+[dependencies]
+rayon.workspace = true
+rustc-hash.workspace = true
+serde.workspace = true
+serde_json.workspace = true
+sqlquack-syntax.workspace = true
+
+[lints]
+workspace = true

crates/sqlquack-lint/Cargo.toml

@@ -0,0 +1,25 @@
+//! Constants shared by the engine modules, and the placeholders hosts write for template code.
+
+pub(crate) const SQL_WILDCARD: &str = "*";
+pub(crate) const OPEN_PAREN: &str = "(";
+pub(crate) const CLOSE_PAREN: &str = ")";
+pub(crate) const EMPTY_FIXTURE_QUERY_TOKEN_COUNT: usize = 6;
+pub(crate) const ENCLOSING_CTE_TOKEN_COUNT: usize = 3;
+/// The indent of each line of a configuration snippet printed in remediation text.
+pub(crate) const SETTING_SNIPPET_INDENT: &str = "            ";
+/// The configuration file remediation text names when the request does not name one.
+pub const DEFAULT_CONFIG_FILE: &str = "sqlclean.toml";
+/// The identifier prefix a host writes in place of each template interpolation.
+pub const INTERPOLATION_PLACEHOLDER_PREFIX: &str = "__sqlclean_interpolation_";
+/// The identifier prefix a host writes in place of a SQL macro parameter.
+pub const MACRO_PARAMETER_PLACEHOLDER_PREFIX: &str = "__sqlclean_macro_parameter_";
+/// The prefix of a relation parameter placeholder, written `<prefix><index>__`.
+pub const RELATION_PARAMETER_PLACEHOLDER_PREFIX: &str = "__sqlclean_relation_parameter_";
+/// The identifier prefix a host writes in place of a value parameter.
+pub const VALUE_PARAMETER_PLACEHOLDER_PREFIX: &str = "__sqlclean_value_parameter_";
+/// Placeholders standing for names the host generates, which Rules never judge as authored names.
+pub(crate) const GENERATED_PLACEHOLDER_PREFIXES: [&str; 3] = [
+    INTERPOLATION_PLACEHOLDER_PREFIX,
+    RELATION_PARAMETER_PLACEHOLDER_PREFIX,
+    VALUE_PARAMETER_PLACEHOLDER_PREFIX,
+];

crates/sqlquack-lint/src/constants.rs

@@ -0,0 +1,18 @@
+//! SQL lint Rules over SQLQuack's tokens and syntax trees, ported from SQLClean's lint engine.
+
+#![forbid(unsafe_code)]
+
+pub mod constants;
+pub mod lint;
+pub mod quality;
+pub(crate) mod tree;
+
+pub use crate::lint::main::batch_engine::lint_batch_json;
+pub use crate::lint::main::catalogue::catalogue;
+pub use crate::lint::main::cross_join_subject::cross_join_subject;
+pub use crate::lint::main::default_rules::default_rules;
+pub use crate::lint::main::engine::lint_json;
+pub use crate::lint::main::fill_finding_path::fill_finding_path;
+pub use crate::lint::main::lint::lint;
+pub use crate::lint::main::parse_sql::parse_sql;
+pub use crate::lint::main::validate_dialect::validate_dialect;