feat: supports a.b.c variables and refactor lexer/parser

[killme2008]

Close #565
Close #529

• New: Regular variables now support nested property/index access without #, e.g. foo.bars[0].name.
• Lookup behavior clarified: For dotted paths like a.b.c, Aviator first checks whether "a.b.c" exists as a full key in the environment; if not, it falls back to chained property traversal (a -> b -> c).
• Stricter syntax validation: Malformed property chains with empty segments (for example a..b) are now rejected with a syntax error.
• Literal safety: Literal keywords (true, false, nil) are consistently treated as literals and cannot be used as indexable values in expressions like true[0].

[Copilot]

Pull request overview

This PR enhances AviatorScript's variable syntax by enabling direct property access notation (e.g., a.b.c) without requiring the # quote prefix, and performs a comprehensive refactoring of the lexer and parser for improved clarity and maintainability.

Key Changes:

• Added support for dotted property access notation (a.b.c) and array index syntax (foo.bars[0].name) without the # prefix
• Refactored parser methods with descriptive names (e.g., ternary() → parseTernary(), expr() → parseAdditive())
• Restructured lexer scanning logic into separate methods (scanNumber(), scanVariable(), scanString(), etc.) for better modularity

Reviewed changes

Copilot reviewed 8 out of 8 changed files in this pull request and generated no comments.

Show a summary per file

File	Description
QuoteVarTest.java	Added tests for property access without quote prefix, including priority behavior when both full key and property chain exist
ExpressionLexerUnitTest.java	Added lexer unit tests for scanning dotted variables with array indices
ExpressionParser.java	Renamed parser methods to be more descriptive, added comprehensive operator precedence documentation, added memory limit for token history, fixed typos
Variable.java	Added utility methods isKeyword() and isLiteralKeyword() to identify reserved variable tokens
OperatorType.java	Enhanced documentation with operator categorization and clarifying comments about token semantics
DelegateToken.java	Fixed typo: Ternay_End → Ternary_End
ExpressionLexer.java	Refactored monolithic scan method into separate scanner methods per token type, improved documentation, optimized character lookup using String.indexOf()
OptimizeCodeGenerator.java	Fixed typo: Ternay_End → Ternary_End

---

💡 Add Copilot custom instructions for smarter, more guided reviews. Learn how to get started.