A8: 文案按A/B/C三套正交叙事生成,避免套路化重复

- constants: 新增 TEXT_NARRATIVE_BY_STRATEGY(A痛点/B场景/C成分),与图片侧同轴
- build_prompt: 加 strategy_narrative 参数并注入 prompt
- text_variants: 全链路透传(含优化轮)
- run_text_generation: 改循环三套,text_count均摊(divmod余前补),跨套去重,打_strategy标记
- TextCandidate: 加 strategy String(4) 字段 + 迁移021(已upgrade head)
- packaging: 打包按strategy精准配对文图(texts_by_strategy映射+三层兜底)
- SSE text_candidate 事件携带 strategy

独立agent交叉验证7改造点全过,边界(text_count<3/无别名/不截断)无must-fix

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

backend/app/models/task.py

@@ -84,6 +84,7 @@ class TextCandidate(Base):
         default=CandidateSource.AI, nullable=False,
     )
     angle_label: Mapped[str | None] = mapped_column(String(64))
+    strategy: Mapped[str | None] = mapped_column(String(4))  # A/B/C 正交叙事套(A痛点/B场景/C成分),与图片侧同轴
     content: Mapped[str | None] = mapped_column(Text)
     score_json: Mapped[str | None] = mapped_column(Text)    # 五维分 JSON
     banned_word_status: Mapped[str] = mapped_column(