feat(summarization): LLM-based code summaries via local OpenAI

[lexasub]

Description

Add LLM-powered code summarization feature (SummarizerService) that generates structured summaries of code entities (functions, methods, classes) with support for Ollama/vLLM backends.

Related Issue

Fixes #(issue number)

Type of Change

• New feature (non-breaking change that adds functionality)

Checklist

• My code follows the code style of this project
• I have added tests that prove my fix/feature works
• All new and existing tests passed (pytest tests/ -v)
• I have updated the documentation accordingly
• My changes generate no new warnings

Testing

# Unit tests
pytest tests/test_summarizer.py -v

# Quality evaluation
ast-rag evaluate --all

# CLI test
ast-rag summarize <function_name> --format markdown

Key Features

• SummarizerService — Ollama/vLLM integration (OpenAI-compatible API)
• Structured output: summary, inputs, outputs, side_effects, calls/called_by, complexity, tags
• CLI: ast-rag summarize <name> --format {markdown,json}
• MCP tool: summarize_code for agent workflows
• Caching: SHA-256 code hash for cache invalidation
• 17 unit tests covering core functionality and edge cases

Configuration

{
  "summarizer": {
    "llm": {
      "base_url": "http://192.168.2.109:1113/v1",
      "model": "qwen2.5-coder:14b"
    },
    "cache": { "enabled": true }
  }
}