feat: Implement full conversion pipeline (PDF/HWP/HWPX/HML/HTML)

- convert.py: 통합 CLI, --json 출력, --scan 폴더 모드
- converters/pdf.py: 페이지별 분류(text/diagram/mixed) + marker-pdf + PNG 렌더링
- converters/hwp.py: COM 자동화 + pyhwp fallback
- converters/hwpx.py: ZIP+XML 직접 파싱, 이미지 추출
- converters/hml.py: XML 파싱, Base64 이미지 추출, colspan/rowspan HTML 표
- converters/html.py: html2text (body_width=0)
- requirements.txt: 최소 의존성
- .env.example: 환경변수 템플릿

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

converters/html.py

@@ -0,0 +1,31 @@
+#!/usr/bin/env python3
+"""HTML / HTM → Markdown (html2text, body_width=0)"""
+from __future__ import annotations
+
+from pathlib import Path
+
+
+def convert_html(html_path: Path, output_dir: Path) -> dict:
+    """HTML → MD. AGENT_GUIDE 스펙 dict 반환."""
+    html_path = Path(html_path)
+    output_dir = Path(output_dir)
+    output_dir.mkdir(parents=True, exist_ok=True)
+    md_path = output_dir / f'{html_path.stem}.md'
+
+    result = {
+        "status": "ok", "input": str(html_path),
+        "output": str(md_path), "format": "html",
+    }
+    try:
+        import html2text
+        h = html2text.HTML2Text()
+        h.body_width = 0
+        h.ignore_links = False
+        h.ignore_images = False
+        content = html_path.read_text(encoding='utf-8', errors='ignore')
+        md = h.handle(content)
+        md_path.write_text(md, encoding='utf-8')
+    except Exception as e:
+        result['status'] = 'error'
+        result['error'] = str(e)
+    return result