wip: phase_z2 evidence 파이프라인 + matching 실험(phase2~26) + 프론트 trace 패널 진행분 스냅샷

- src: phase_z2 composition/mapper/pipeline/placement_planner/retry, ai_fallback(prompts/schema/validate), mdx_text_atoms 신규
- Front: PipelineTracePanel 신규, FramePanel/SlideCanvas/Home/designAgentApi 등 갱신 + 테스트 4종 추가
- templates/phase_z2: catalog(component_expansion_registry, node_slot_mapping 신규), frames, families, slide_base 갱신
- tests/matching: phase2~26 매칭 실험 스크립트·리포트·온톨로지 전체 (미커밋 진행분)
- tests: b4_v4 evidence, task5~28.5 시리즈, regression(imp95 baseline) 등 신규 테스트 대량 추가
- docs/reference: MDX 구조 인벤토리, MDX→Frame 구조 계약 문서
- scripts: mdx 계약/parity/coverage/viewport 체크, gitea comment, run sync 유틸
- .gitignore: tmp*.json, chromedriver, .orchestrator, *.pkl, Front_test* 등 임시/스냅샷 제외

미완성 작업의 보존용 스냅샷 커밋 (2026-07-02)

Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>

tests/test_mdx_text_atoms.py

@@ -0,0 +1,56 @@
+from pathlib import Path
+
+from src.mdx_text_atoms import compare_atom_sets, extract_text_atoms, normalize_text_atom
+
+
+def test_extract_atoms_ignores_syntax_and_keeps_semantic_text():
+    mdx = """---
+title: demo
+---
+
+import Demo from './Demo.astro'
+
+## 1. Title
+
+- **Important** text
+<details>
+  <summary style={{cursor: 'pointer'}}>More info</summary>
+  <div style={{ color: '#999' }}>
+    <p>Nested semantic text</p>
+  </div>
+</details>
+
+| A | B |
+| :--- | :--- |
+| one | two |
+
+![Diagram caption](/assets/demo.png)
+"""
+    atoms = extract_text_atoms(mdx, source_name="demo.mdx")
+    texts = [a.normalized for a in atoms]
+
+    assert "1. Title" in texts
+    assert "Important text" in texts
+    assert "More info" in texts
+    assert "Nested semantic text" in texts
+    assert "A" in texts
+    assert "B" in texts
+    assert "one" in texts
+    assert "two" in texts
+    assert "Diagram caption" in texts
+    assert not any("cursor" in t for t in texts)
+
+
+def test_compare_atom_sets_reports_missing_and_added():
+    original = extract_text_atoms("- Original text\n- Shared text\n", source_name="original")
+    standardized = extract_text_atoms("- Shared text\n- Added text\n", source_name="standardized")
+
+    result = compare_atom_sets(original, standardized)
+
+    assert [a.normalized for a in result["missing_from_standardized"]] == ["Original text"]
+    assert [a.normalized for a in result["added_in_standardized"]] == ["Added text"]
+
+
+def test_normalize_text_atom_removes_markdown_markers_only():
+    assert normalize_text_atom("**Digital** `Transformation`") == "Digital Transformation"
+