feat(question_gen): AnchorContext + 12 题型-层级映射常量

- AnchorContext frozen dataclass: 锚节点生成上下文（node_id, card_text, frame_paths, subtitle, distractor_texts）
- TaskTypeSpec frozen dataclass: 题型生成规格（level, needs_frames, frame_count, context_fields）
- TASK_TYPE_LEVEL_MAP: 12 种 Video-MME 题型 → 树层级 + 生成规格映射
- 11 项单元测试覆盖：映射完整性、值类型、层级合法性、frozen 不变性

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

app/question_gen/synthesizer.py

@@ -0,0 +1,68 @@
+"""赛题合成核心逻辑 — 节点采样、prompt 构造、去重。
+
+纯函数为主，异步编排仅 generate_one。
+通过 DI 接收 VLMProvider / EmbeddingProvider，不 import adapters/。
+"""
+
+from __future__ import annotations
+
+from dataclasses import dataclass
+
+
+@dataclass(frozen=True)
+class AnchorContext:
+    """锚节点上下文——生成单道题所需的全部素材。
+
+    属性:
+        node_id: 锚节点 ID。
+        card_text: 锚节点 card 序列化文本。
+        frame_paths: 帧图片路径列表。
+        subtitle: 对应字幕（可空）。
+        distractor_texts: 同视频其他节点摘要（供 VLM 生成干扰项）。
+    """
+
+    node_id: str
+    card_text: str
+    frame_paths: list[str]
+    subtitle: str
+    distractor_texts: list[str]
+
+
+@dataclass(frozen=True)
+class TaskTypeSpec:
+    """题型的生成规格。
+
+    属性:
+        level: 锚定层级（"L3" / "L2" / "L1" / "L1-L2"）。
+        needs_frames: 是否必须提供帧图。
+        frame_count: 帧数范围描述（如 "1", "2-3", "0-1"）。
+        context_fields: 需要提取的 card 字段元组。
+    """
+
+    level: str
+    needs_frames: bool
+    frame_count: str
+    context_fields: tuple[str, ...]
+
+
+# ---------------------------------------------------------------------------
+# 12 种 Video-MME 题型 → 树层级 + 生成规格映射
+# ---------------------------------------------------------------------------
+
+TASK_TYPE_LEVEL_MAP: dict[str, TaskTypeSpec] = {
+    # --- L3 单帧题型 ---
+    "Object Recognition": TaskTypeSpec("L3", True, "1", ("frame_summary",)),
+    "Attribute Perception": TaskTypeSpec("L3", True, "1", ("frame_summary",)),
+    "OCR Problems": TaskTypeSpec("L3", True, "1", ("frame_summary",)),
+    "Spatial Reasoning": TaskTypeSpec("L3", True, "1", ("frame_summary", "spatial_layout")),
+    "Spatial Perception": TaskTypeSpec("L3", True, "1", ("frame_summary",)),
+    # --- L2 多帧 / 事件级题型 ---
+    "Action Recognition": TaskTypeSpec("L2", True, "2-3", ("event_description",)),
+    "Action Reasoning": TaskTypeSpec("L2", True, "2-3", ("event_description",)),
+    "Counting Problem": TaskTypeSpec("L2", True, "2-3", ("event_description",)),
+    "Temporal Perception": TaskTypeSpec("L2", False, "0-1", ("event_description", "time_range")),
+    # --- L1 / 跨层级题型 ---
+    "Temporal Reasoning": TaskTypeSpec("L1", True, "per-L2", ("scene_summary",)),
+    "Information Synopsis": TaskTypeSpec("L1", True, "per-L2", ("scene_summary",)),
+    "Object Reasoning": TaskTypeSpec("L1-L2", True, "per-L2", ("event_description",)),
+}

tests/unit/test_synthesizer.py

@@ -0,0 +1,134 @@
+"""synthesizer 模块单元测试 — AnchorContext + 题型映射常量。"""
+
+from __future__ import annotations
+
+import dataclasses
+
+from app.question_gen.synthesizer import TASK_TYPE_LEVEL_MAP, AnchorContext, TaskTypeSpec
+
+ALL_12_TYPES = [
+    "Object Recognition",
+    "Attribute Perception",
+    "OCR Problems",
+    "Spatial Reasoning",
+    "Spatial Perception",
+    "Action Recognition",
+    "Action Reasoning",
+    "Counting Problem",
+    "Temporal Perception",
+    "Temporal Reasoning",
+    "Information Synopsis",
+    "Object Reasoning",
+]
+
+
+class TestTaskTypeLevelMap:
+    """TASK_TYPE_LEVEL_MAP 覆盖性与结构测试。"""
+
+    def test_covers_all_12_types(self) -> None:
+        """映射表必须覆盖全部 12 种 Video-MME 题型。"""
+        assert set(TASK_TYPE_LEVEL_MAP.keys()) == set(ALL_12_TYPES)
+
+    def test_no_extra_types(self) -> None:
+        """映射表不得包含 12 种标准题型之外的条目。"""
+        assert len(TASK_TYPE_LEVEL_MAP) == 12
+
+    def test_all_values_are_task_type_spec(self) -> None:
+        """每个映射值必须是 TaskTypeSpec 实例。"""
+        for task_type, spec in TASK_TYPE_LEVEL_MAP.items():
+            assert isinstance(spec, TaskTypeSpec), f"{task_type} 映射值类型错误: {type(spec)}"
+
+    def test_level_values_valid(self) -> None:
+        """每个 spec 的 level 必须是合法层级标识。"""
+        valid_levels = {"L1", "L2", "L3", "L1-L2"}
+        for task_type, spec in TASK_TYPE_LEVEL_MAP.items():
+            assert spec.level in valid_levels, (
+                f"{task_type} 层级 '{spec.level}' 不在 {valid_levels}"
+            )
+
+    def test_context_fields_non_empty(self) -> None:
+        """每个 spec 的 context_fields 至少有一个字段。"""
+        for task_type, spec in TASK_TYPE_LEVEL_MAP.items():
+            assert len(spec.context_fields) >= 1, f"{task_type} 的 context_fields 为空"
+
+
+class TestAnchorContext:
+    """AnchorContext 数据类测试。"""
+
+    def test_frozen(self) -> None:
+        """AnchorContext 是不可变的。"""
+        ctx = AnchorContext(
+            node_id="L3_001",
+            card_text="A person walks into a room",
+            frame_paths=["/data/frames/001.jpg"],
+            subtitle="Hello there",
+            distractor_texts=["A car drives by"],
+        )
+        assert ctx.node_id == "L3_001"
+        assert ctx.card_text == "A person walks into a room"
+        assert ctx.frame_paths == ["/data/frames/001.jpg"]
+        assert ctx.subtitle == "Hello there"
+        assert ctx.distractor_texts == ["A car drives by"]
+
+    def test_mutation_raises(self) -> None:
+        """frozen dataclass 拒绝赋值修改。"""
+        ctx = AnchorContext(
+            node_id="L3_001",
+            card_text="test",
+            frame_paths=["a.jpg"],
+            subtitle="",
+            distractor_texts=["other node"],
+        )
+        try:
+            ctx.node_id = "L3_002"  # type: ignore[misc]
+            raise AssertionError("应抛出 FrozenInstanceError")
+        except dataclasses.FrozenInstanceError:
+            pass
+
+    def test_empty_subtitle_allowed(self) -> None:
+        """subtitle 可以为空字符串。"""
+        ctx = AnchorContext(
+            node_id="L2_010",
+            card_text="scene card",
+            frame_paths=[],
+            subtitle="",
+            distractor_texts=[],
+        )
+        assert ctx.subtitle == ""
+
+    def test_multiple_frame_paths(self) -> None:
+        """frame_paths 可包含多个路径。"""
+        paths = ["/data/f1.jpg", "/data/f2.jpg", "/data/f3.jpg"]
+        ctx = AnchorContext(
+            node_id="L2_005",
+            card_text="multi-frame event",
+            frame_paths=paths,
+            subtitle="Dialogue line",
+            distractor_texts=["other1", "other2"],
+        )
+        assert len(ctx.frame_paths) == 3
+
+
+class TestTaskTypeSpec:
+    """TaskTypeSpec 数据类测试。"""
+
+    def test_frozen(self) -> None:
+        """TaskTypeSpec 是不可变的。"""
+        spec = TaskTypeSpec(
+            level="L3",
+            needs_frames=True,
+            frame_count="1",
+            context_fields=("frame_summary",),
+        )
+        try:
+            spec.level = "L2"  # type: ignore[misc]
+            raise AssertionError("应抛出 FrozenInstanceError")
+        except dataclasses.FrozenInstanceError:
+            pass
+
+    def test_context_fields_is_tuple(self) -> None:
+        """context_fields 应为 tuple（不可变）。"""
+        for task_type, spec in TASK_TYPE_LEVEL_MAP.items():
+            assert isinstance(spec.context_fields, tuple), (
+                f"{task_type} 的 context_fields 不是 tuple"
+            )