Fix Top Songs: accumulate play counts per scrape, deduplicate only stat card

song_counts tallies how many times each song appears across all runs
(powers the Top Songs ranking). total_scraped uses the unique seen
set separately so the stat card stays deduplicated.

Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>

journeyfm/web_service.py

@@ -68,9 +68,8 @@ def load_recent_stats(db_path=None):
                     sname = song.get("source", "Unknown")
                     title = song.get("title", "?")
                     artist = song.get("artist", "?")
-                    k = (artist, title)
-                    if k not in scraped_seen:
-                        scraped_seen.add(k)
+                    scraped_seen.add((artist, title))
+                    # accumulate play count every scrape (powers Top Songs ranking)
                     stats["song_counts"].setdefault(sname, {})
                     dk = f"{artist} - {title}"
                     stats["song_counts"][sname][dk] = stats["song_counts"][sname].get(dk, 0) + 1