feat: support string query functions

[dodok8]

Summary

Add initial query-language support for built-in string value functions.

Closes #17

Scope

• Documented the MVP string function contract for concat(...) and str(expr).
• Added parser coverage for variadic concat and str function-call syntax preservation.
• Added StringFunctionExpr to Semantic IR with argument scalar type and cardinality metadata.
• Resolved concat(...) and str(expr) with resolver-level arity, type, null, and many-cardinality checks.
• Lowered resolved string functions into SQLite query plans without exposing SQLite operators in query syntax.
• Rendered concat as SQLite || and str(bool) as a null-aware CASE expression returning true/false text.
• Added query-pipeline render and execution coverage for string functions.

Spec and plan alignment

List the spec/ and plan/ files checked for this change. If the change
intentionally diverges from them, explain why and include the required document
update in this pull request.

• spec/query.md
• spec/ir.md
• spec/sqlite-query-plan.md
• plan/query-expression-model-plan.md
• No intentional divergence. The spec updates are included in this pull request.

Tests

List the automated and manual checks performed.

• cargo test --workspace
• Other: targeted crate tests were run during TDD for query-parser, query-ir, query-resolver, sqlite-query-plan, sqlite-query-sqlgen, and query-pipeline-tests.

Review notes

Call out API boundaries, temporary rules, migration concerns, or follow-up work
that reviewers should inspect closely.

• concat(...) is variadic but requires at least two str arguments.
• str(expr) is explicit and does not introduce implicit string casts for comparisons, membership checks, arithmetic, or concatenation.
• str(bool) intentionally renders true/false text through SQLite CASE instead of relying on SQLite CAST, which would expose 1/0 text behavior.
• Other string helpers such as length, lower, upper, trim, replace, substr, and predicate helpers remain deferred.

AI assistance

Disclose AI tool usage according to AI_POLICY.md.

• No AI tools were used.
• AI tools were used and are disclosed here and in the relevant commit
  messages with Assisted-by trailers.

Tools and models:

• Codex:gpt-5.5

[dodok8]

@codex review

[chatgpt-codex-connector]

Codex Review: Didn't find any major issues. Keep it up!

Reviewed commit: 65824e07db

ℹ️ About Codex in GitHub

Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you

• Open a pull request for review
• Mark a draft as ready
• Comment "@codex review".

If Codex has suggestions, it will comment; otherwise it will react with 👍.

Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".

[sonarqubecloud]

Quality Gate passed

Issues
0 New issues
0 Accepted issues

Measures
0 Security Hotspots
96.4% Coverage on New Code
1.0% Duplication on New Code

See analysis details on SonarQube Cloud