sync: update all skills from latest workspace code

doc_parser_skill:
- New: verify_flowchart.py (flowchart validation)
- Updated: LLM.py (multi-provider: DeepSeek + DashScope)
- Updated: image_parser.py (logic tree support, external prompts)
- Updated: SKILL.md, prompts/image_prompt.md

conflict_detection_skill:
- Updated: LLM.py (multi-provider sync)
- Updated: detect_conflicts.py (logic tree text conversion)

ir_generation_skill:
- Replaced old scripts/LLM.py + ir_generator.py with standalone project
- New: main.py, config.py, step1-3_*.py, ensemble_merge.py
- New: prompts/, tests/ subdirectories

tests:
- New: acceptance/ test suite with schema validation
- Fixed: conftest no longer globally skips non-acceptance tests
- Updated: test_sample.py for new ir_generation structure

Co-Authored-By: Claude Opus 4.7 <noreply@anthropic.com>

skills/ir_generation_skill/tests/test_step2_5.py

@@ -0,0 +1,152 @@
+"""
+Tests for Stage 2.5 (Branch Coverage Auto-Completion).
+
+Validates:
+- Path enumeration exists and is non-empty
+- Auto-complete fragments have valid structure
+- No duplicate unit_ids in autocomplete fragments
+- Path coverage improved after autocomplete (if applicable)
+"""
+
+import sys
+from pathlib import Path
+
+sys.path.insert(0, str(Path(__file__).parent.parent))
+import config
+
+
+PASS = "[PASS]"
+FAIL = "[FAIL]"
+WARN = "[WARN]"
+
+
+def load_path_enumeration():
+    """Load path_enumeration.json."""
+    try:
+        return config.load_json(config.PATH_ENUM_JSON)
+    except FileNotFoundError:
+        print(f"{FAIL} path_enumeration.json 未找到: {config.PATH_ENUM_JSON}")
+        print("  请先运行 step2_5_branch_coverage.py")
+        sys.exit(1)
+
+
+def load_autocomplete_fragments():
+    """Load ir_autocomplete_fragments.json, or return [] if absent."""
+    path = config.IR_AUTOCOMPLETE_FRAGMENTS_JSON
+    if not Path(path).exists():
+        return None
+    return config.load_json(path)
+
+
+def check_path_enumeration(data: dict) -> list[str]:
+    """Check path enumeration has valid structure."""
+    errors = []
+    paths = data.get("logic_tree_paths", {})
+    if not paths:
+        errors.append("logic_tree_paths 为空")
+    total = data.get("total_paths", 0)
+    if total <= 0:
+        errors.append(f"total_paths = {total}, 期望 > 0")
+
+    for image_id, image_paths in paths.items():
+        if not image_paths:
+            errors.append(f"{image_id}: 路径列表为空")
+            continue
+        for i, p in enumerate(image_paths):
+            if not p.get("path_id"):
+                errors.append(f"{image_id}[{i}]: 缺少 path_id")
+            if not p.get("image_id"):
+                errors.append(f"{image_id}[{i}]: 缺少 image_id")
+            if not p.get("node_ids"):
+                errors.append(f"{image_id}[{i}]: 缺少 node_ids")
+
+    return errors
+
+
+def check_autocomplete_fragments(fragments: list[dict] | None) -> list[str]:
+    """Check auto-complete fragments have valid structure."""
+    if fragments is None:
+        return ["ir_autocomplete_fragments.json 未生成 (可能无需补全)"]
+
+    errors = []
+    seen_unit_ids = set()
+
+    for frag in fragments:
+        uid = frag.get("unit_id", "")
+        if not uid:
+            errors.append("fragment 缺少 unit_id")
+            continue
+        if uid in seen_unit_ids:
+            errors.append(f"unit_id '{uid}' 重复")
+        seen_unit_ids.add(uid)
+
+        if not frag.get("auto_generated"):
+            errors.append(f"{uid}: auto_generated 应为 true")
+
+        rules = frag.get("rules", [])
+        for j, rule in enumerate(rules):
+            rid = rule.get("rule_id", f"rule[{j}]")
+            if not rule.get("path"):
+                errors.append(f"{rid}: path 字段缺失")
+            precond = rule.get("precondition", {})
+            if not precond.get("geographic_scope"):
+                errors.append(f"{rid}: precondition.geographic_scope 缺失")
+
+    return errors
+
+
+def run_all_tests():
+    print("=" * 60)
+    print("Step 2.5 自检测试")
+    print("=" * 60)
+
+    all_errors = []
+
+    # Test 1: Path enumeration exists
+    try:
+        path_data = load_path_enumeration()
+    except SystemExit:
+        return False
+
+    errors = check_path_enumeration(path_data)
+    if errors:
+        print(f"\n{FAIL} 路径枚举检查: {len(errors)} 个错误")
+        for e in errors:
+            print(f"  - {e}")
+        all_errors.extend(errors)
+    else:
+        total = path_data.get("total_paths", 0)
+        n_images = len(path_data.get("logic_tree_paths", {}))
+        print(f"\n{PASS} 路径枚举检查: {total} 条路径, {n_images} 个逻辑树")
+
+    # Test 2: Auto-complete fragments
+    fragments = load_autocomplete_fragments()
+    errors = check_autocomplete_fragments(fragments)
+
+    if fragments is None:
+        print(f"\n{WARN} 自动补全片段: 未生成 (可能所有路径已覆盖)")
+    elif errors:
+        print(f"\n{FAIL} 自动补全片段检查: {len(errors)} 个错误")
+        for e in errors[:10]:
+            print(f"  - {e}")
+        all_errors.extend(errors)
+    else:
+        auto_rules = sum(len(f.get("rules", [])) for f in fragments)
+        print(f"\n{PASS} 自动补全片段检查: "
+              f"{len(fragments)} 个片段, {auto_rules} 条规则")
+
+    # Summary
+    print(f"\n{'='*60}")
+    total_failures = len(all_errors)
+
+    if total_failures == 0:
+        print(f"{PASS} 所有测试通过!")
+    else:
+        print(f"{FAIL} 测试失败: {total_failures} 个错误")
+
+    return total_failures == 0
+
+
+if __name__ == "__main__":
+    success = run_all_tests()
+    sys.exit(0 if success else 1)