重构回忆录为 story-first / markdown-first 架构并整合图片意图与前端 UI 修复

本次 squash merge 将 codex-story-first-image-intent 的整体改动合入 development，核心内容包括：

1. 后端数据与迁移：新增 stories、story_versions、story_image_intents、chapter_cover_intents、assets 等模型与 Alembic 迁移，建立 story-first、markdown-first、asset-first 的主数据链路。

2. 生成与任务链：引入 StoryBuilderOrchestrator、ChapterComposerOrchestrator、story_image_tasks、chapter_cover_tasks，图片生成从正文占位符改为结构化 intent -> asset -> markdown 回填。

3. 并发与一致性：为 story/chapter intent 增加 claim_token、claimed_at、attempt_count，采用数据库原子 claim 为主、Redis 锁为辅，避免重复生成、锁误删和 processing 卡死。

4. Memoir 读写路径：章节 canonical_markdown 成为正文真源，列表/详情接口补齐 markdown、cover_asset、word_count 等字段，PDF 与 asset 解析链路同步升级。

5. Memory / Retrieval：扩展 transcript ingest、chunking、evidence 检索与 story 聚合基础设施，为后续 story-first RAG 与多 agent 编排提供底座。

6. App 端体验：章节页继续走 MarkdownRenderer 阅读链，同时吸收 fix3-19 的跨平台 UI glitch 修复；更新对话页、首页、文案资源与章节列表映射逻辑。

7. 测试与文档：补充 asset resolver、story image task、章节封面派发、markdown 映射等回归测试，并加入图片占位符退役设计文档。

api/app/features/memoir/pdf_service.py

@@ -4,7 +4,7 @@ PDF 生成服务（从 services 迁入 memoir feature）
 
 from app.core.logging import get_logger
 from io import BytesIO
-from typing import List
+from typing import List, Optional
 
 import httpx
 from PIL import Image
@@ -21,12 +21,20 @@ from reportlab.platypus import (
 from reportlab.pdfbase import pdfmetrics
 from reportlab.pdfbase.cidfonts import UnicodeCIDFont
 
+from app.features.memoir.asset_resolver import (
+    collect_asset_ids_from_markdown,
+    split_markdown_by_asset_refs,
+    strip_legacy_image_placeholders,
+)
+from app.features.memoir.helpers import (
+    _chapter_markdown,
+    sections_to_content_and_images,
+)
 from app.features.memoir.memoir_images.parser import PLACEHOLDER_RE
 from app.features.memoir.memoir_images.schema import (
     IMAGE_STATUS_COMPLETED,
     normalize_image_assets,
 )
-from app.features.memoir.memoir_images.serializers import memoir_image_to_dict
 from app.features.memoir.memoir_images.storage import (
     CosDownloadUrlError,
     TencentCosStorageService,
@@ -60,24 +68,6 @@ def split_content_blocks(content: str, images: list[dict]) -> list[dict]:
     return blocks
 
 
-def sections_to_blocks(sections: list, prepare_fn=None) -> list[dict]:
-    if prepare_fn is None:
-        prepare_fn = _prepare_pdf_image_assets
-    blocks: list[dict] = []
-    for section in sorted(sections, key=lambda s: getattr(s, "order_index", 0)):
-        content = (getattr(section, "content", None) or "").strip()
-        if content:
-            blocks.append({"type": "text", "value": content})
-        img = None
-        if getattr(section, "image_record", None):
-            img = memoir_image_to_dict(section.image_record)
-        if img:
-            prepared = prepare_fn([img])
-            if prepared and prepared[0].get("url"):
-                blocks.append({"type": "image", "url": prepared[0]["url"]})
-    return blocks
-
-
 def _prepare_pdf_image_assets(images: list[dict]) -> list[dict]:
     storage = TencentCosStorageService.from_env()
     prepared_assets: list[dict] = []
@@ -132,7 +122,12 @@ class PDFService:
             logger.warning("PDF 图片下载失败: url=%s, error=%s", url, exc)
             return None
 
-    async def generate_pdf(self, book, chapters: List) -> bytes:
+    async def generate_pdf(
+        self,
+        book,
+        chapters: List,
+        asset_url_map: Optional[dict[str, str]] = None,
+    ) -> bytes:
         buffer = BytesIO()
         doc = SimpleDocTemplate(buffer, pagesize=A4)
         styles = getSampleStyleSheet()
@@ -170,16 +165,27 @@ class PDFService:
         for chapter in chapters:
             story.append(Paragraph(chapter.title, heading_style))
             story.append(Spacer(1, 0.2 * inch))
-            sections = getattr(chapter, "sections", None) or []
-            if sections:
-                blocks = sections_to_blocks(sections)
+            # 正文真源：canonical_markdown（与 API / 前端一致）
+            markdown = _chapter_markdown(chapter)
+            _, images_list = sections_to_content_and_images(chapter)
+            if not markdown:
+                markdown = getattr(chapter, "content", "") or ""
+            if not images_list:
+                images_list = list(getattr(chapter, "images", None) or [])
+            prepared_images = _prepare_pdf_image_assets(images_list)
+            blocks: list[dict]
+            if asset_url_map and collect_asset_ids_from_markdown(markdown):
+                blocks = split_markdown_by_asset_refs(
+                    markdown,
+                    lambda aid: asset_url_map.get(aid) if asset_url_map else None,
+                )
+                for b in blocks:
+                    if b.get("type") == "text":
+                        b["value"] = strip_legacy_image_placeholders(
+                            b.get("value") or ""
+                        )
             else:
-                images = _prepare_pdf_image_assets(
-                    getattr(chapter, "images", None) or []
-                )
-                blocks = split_content_blocks(
-                    getattr(chapter, "content", "") or "", images
-                )
+                blocks = split_content_blocks(markdown, prepared_images)
             for block in blocks:
                 if block["type"] == "text":
                     paragraphs = block["value"].split("\n\n")