refactor(ranking): replace blob cache with per-item PocketBase storage

- Replace SetRanking/GetRanking/SetRankingPageHTML/GetRankingPageHTML blob methods
  with WriteRankingItem/ReadRankingItems/RankingFreshEnough per-item operations
- Add 24h staleness gate in ScrapeRanking to skip re-scraping fresh data
- Add GET /api/ranking endpoint returning []RankingItem sorted by rank
- Remove RankingPageCacher interface and rankingCacheAdapter adapter
- Update integration tests to use new per-item upsert semantics
- Include e2e test suite (scraper/internal/e2e/)

scraper/internal/scraper/interfaces.go

@@ -120,16 +120,6 @@ type RankingProvider interface {
 	ScrapeRanking(ctx context.Context, maxPages int) (<-chan BookMeta, <-chan error)
 }
 
-// RankingPageCacher persists and retrieves raw HTML for individual ranking pages.
-// Implementations (e.g. writer.Writer) store files on disk so that a
-// subsequent ScrapeRanking call can serve cached HTML without a network round-trip.
-type RankingPageCacher interface {
-	// WriteRankingPageCache stores the raw HTML string for the given page number.
-	WriteRankingPageCache(page int, html string) error
-	// ReadRankingPageCache returns the cached HTML for page, or ("", nil) on a miss.
-	ReadRankingPageCache(page int) (string, error)
-}
-
 // NovelScraper is the full interface that a concrete novel source must implement.
 // It composes all four provider interfaces.
 type NovelScraper interface {