Scraper: don't flag not_listed from partial scrapes

A page-load timeout was logged and skipped, so a scrape could 'succeed'
with only page 1 of results — and the not_listed flagging then marked
every page-2 hotel absent, suppressing their last known rates.

- ScraperResult now tracks pages_requested/pages_ok
- scrape_date skips not_listed flagging when pages failed or when the
  scrape saw <60% of the date's 7-day coverage baseline; scraped rates
  are still saved, unseen hotels keep last known rate + scrape time

Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>

backend/services/scraper_backends/base.py

@@ -59,6 +59,8 @@ class ScraperResult:
     rates: List[RateData] = field(default_factory=list)
     error_message: Optional[str] = None
     page_content_sample: Optional[str] = None  # For debugging
+    pages_requested: int = 0             # Result pages we set out to fetch
+    pages_ok: int = 0                    # Pages that loaded and parsed cleanly
 
 
 class ScraperBackend(ABC):

backend/services/scraper_backends/playwright_local.py

@@ -292,6 +292,7 @@ class PlaywrightLocalBackend(ScraperBackend):
         all_hotels = []
         all_rates = []
         seen_hotel_ids = set()
+        pages_ok = 0
 
         context = None
         page = None
@@ -315,10 +316,12 @@ class PlaywrightLocalBackend(ScraperBackend):
 
                 logger.info(f"Scraping page {page_num + 1}: {url}")
 
+                page_loaded = True
                 try:
                     await page.goto(url, wait_until='networkidle', timeout=30000)
                 except Exception as e:
                     logger.warning(f"Page load timeout, continuing: {e}")
+                    page_loaded = False
 
                 # Check for blocking
                 content = await page.content()
@@ -331,7 +334,9 @@ class PlaywrightLocalBackend(ScraperBackend):
                         block_reason=reason,
                         hotels=all_hotels,
                         rates=all_rates,
-                        page_content_sample=content[:1000]
+                        page_content_sample=content[:1000],
+                        pages_requested=pages,
+                        pages_ok=pages_ok,
                     )
 
                 # Human-like scrolling
@@ -340,6 +345,12 @@ class PlaywrightLocalBackend(ScraperBackend):
                 # Extract data
                 hotels, rates = await self._extract_search_results(page, check_in)
 
+                # A page counts as clean if it loaded fully and parsed.
+                # An empty page 1 means the results never rendered; an empty
+                # later page can legitimately be the end of the results.
+                if page_loaded and (hotels or page_num > 0):
+                    pages_ok += 1
+
                 # Deduplicate by booking_com_id
                 for hotel, rate in zip(hotels, rates):
                     if hotel.booking_com_id and hotel.booking_com_id not in seen_hotel_ids:
@@ -353,7 +364,9 @@ class PlaywrightLocalBackend(ScraperBackend):
                 success=True,
                 blocked=False,
                 hotels=all_hotels,
-                rates=all_rates
+                rates=all_rates,
+                pages_requested=pages,
+                pages_ok=pages_ok,
             )
 
         except Exception as e:
@@ -363,7 +376,9 @@ class PlaywrightLocalBackend(ScraperBackend):
                 blocked=False,
                 error_message=str(e),
                 hotels=all_hotels,
-                rates=all_rates
+                rates=all_rates,
+                pages_requested=pages,
+                pages_ok=pages_ok,
             )
         finally:
             if page: