Fix leaderboard scrolling performance after PR #39 merge (#63)

* Fix leaderboard scrolling performance after PR #39 merge

- Restore leaderboard background updates that were accidentally removed
- Fix duration method call from get_dynamic_duration() back to get_duration()
- Restore proper fallback duration (600s instead of 60s) for leaderboard
- Add back sports manager updates that feed data to leaderboard
- Fix leaderboard defer_update priority to prevent scrolling lag

These changes restore the leaderboard's dynamic duration calculation
and ensure it gets proper background updates for smooth scrolling.

* Apply PR #60 leaderboard performance optimizations

- Change scroll_delay from 0.05s to 0.01s (100fps instead of 20fps)
- Remove conditional scrolling logic - scroll every frame for smooth animation
- Add FPS tracking and logging for performance monitoring
- Restore high-framerate scrolling that was working before PR #39 merge

These changes restore the smooth leaderboard scrolling performance
that was achieved in PR #60 but was lost during the PR #39 merge.

* Fix critical bugs identified in PR #39 review

- Fix record filtering logic bug: change away_record == set to away_record in set
- Fix incorrect sport specification: change 'nfl' to 'ncaa_fb' for NCAA Football data requests
- These bugs were causing incorrect data display and wrong sport data fetching

Addresses issues found by cursor bot in PR #39 review:
- Record filtering was always evaluating to False
- NCAA Football was fetching NFL data instead of college football data

* Enhance cache clearing implementation from PR #39

- Add detailed logging to cache clearing process for better visibility
- Log cache clearing statistics (memory entries and file count)
- Improve startup logging to show cache clearing and data refetch process
- Addresses legoguy1000's comment about preventing stale data issues

This enhances the cache clearing implementation that was added in PR #39
to help prevent legacy cache issues and stale data problems.

* continuing on base_classes - added baseball and api extractor since we don't use ESPN api for all sports

* tests

* fix missing duration

* ensure milb, mlb, ncaa bb are all using new baseball base class properly

* cursor rule to help with PR creation

* fix image call

* fix _scoreboard suffix on milb, MLB

src/ncaa_fb_managers.py

@@ -28,6 +28,9 @@ class BaseNCAAFBManager(Football): # Renamed class
     def __init__(self, config: Dict[str, Any], display_manager: DisplayManager, cache_manager: CacheManager):
         self.logger = logging.getLogger('NCAAFB') # Changed logger name
         super().__init__(config=config, display_manager=display_manager, cache_manager=cache_manager, logger=self.logger, sport_key="ncaa_fb")
+        
+        # Configuration is already set in base class
+        # self.logo_dir and self.update_interval are already configured
 
         # Check display modes to determine what data to fetch
         display_modes = self.mode_config.get("display_modes", {})
@@ -140,7 +143,7 @@ class BaseNCAAFBManager(Football): # Renamed class
         
         # Submit background fetch request
         request_id = self.background_service.submit_fetch_request(
-            sport="nfl",
+            sport="ncaa_fb",
             year=season_year,
             url=ESPN_NCAAFB_SCOREBOARD_URL,
             cache_key=cache_key,
@@ -184,7 +187,7 @@ class BaseNCAAFBManager(Football): # Renamed class
         except requests.exceptions.RequestException as e:
             self.logger.error(f"[API error fetching full schedule: {e}")
             return None
-    
+
     def _fetch_data(self) -> Optional[Dict]:
         """Fetch data using shared data mechanism or direct fetch for live."""
         if isinstance(self, NCAAFBLiveManager):