Introduce single-pass AST traversal

[CrazyDubya]

Summary

• walk AST once using _traverse_tree in analyzers
• delegate per-node analysis for types, performance, memory, dependencies, and complexity
• adapt code generator to use AnalysisResult.type_info
• add early tuple-unpack handling in CodeAnalyzer

Testing

• pip install -q -r requirements.txt
• pytest -q

---

https://chatgpt.com/codex/tasks/task_e_684a55e8aa3083329556c5a77994473d

[Copilot]

Pull Request Overview

This PR refactors the AST analysis process to use a single-pass traversal instead of multiple separate tree walks. The changes convert per-tree analysis methods to per-node analysis and update the code generator to use the new AnalysisResult.type_info structure.

• Introduces _traverse_tree method for single-pass AST traversal with delegated per-node analysis
• Refactors analysis methods to operate on individual nodes instead of entire trees
• Updates code generator to access type information through AnalysisResult.type_info

Reviewed Changes

Copilot reviewed 3 out of 3 changed files in this pull request and generated 5 comments.

File	Description
src/converter/code_generator.py	Updates function declaration/implementation generation to use AnalysisResult.type_info
src/analyzer/code_analyzer_fixed.py	Implements single-pass traversal and refactors analysis methods for per-node processing
src/analyzer/code_analyzer.py	Adds single-pass traversal and extensive tuple unpacking logic in type inference

[Copilot]

The tuple unpacking logic (lines 98-149) is extremely complex and deeply nested, making it difficult to maintain. Consider extracting this into a separate method like _handle_tuple_unpacking(node) to improve readability and testability.