fix(userdata): read user data through to the database

The user data cache and the item's in-memory rows were both filled once per
replica and never invalidated, so a pod serving a playback tick read its own
stale resume position and wrote it back over the position another pod had just
saved, silently losing resume points, played state, favourites and ratings.

- drop the user item data cache
- read single and batched user data from the database on every query
- prefetch user data for in-memory sorts and filters so each stays one query
- cover the lost update against real PostgreSQL with two manager instances

Emby.Server.Implementations/Sorting/IsPlayedComparer.cs

@@ -2,6 +2,8 @@
 
 #pragma warning disable CS1591
 
+using System;
+using System.Collections.Generic;
 using Jellyfin.Data.Enums;
 using Jellyfin.Database.Implementations.Entities;
 using MediaBrowser.Controller.Entities;
@@ -36,6 +38,12 @@ namespace Emby.Server.Implementations.Sorting
         /// <value>The user manager.</value>
         public IUserManager UserManager { get; set; }
 
+        /// <summary>
+        /// Gets or sets the prefetched user data.
+        /// </summary>
+        /// <value>The prefetched user data.</value>
+        public IReadOnlyDictionary<Guid, UserItemData> PrefetchedUserData { get; set; }
+
         /// <summary>
         /// Compares the specified x.
         /// </summary>
@@ -54,7 +62,7 @@ namespace Emby.Server.Implementations.Sorting
         /// <returns>DateTime.</returns>
         private int GetValue(BaseItem x)
         {
-            return x.IsPlayed(User, userItemData: null) ? 0 : 1;
+            return x.IsPlayed(User, this.GetUserData(x)) ? 0 : 1;
         }
     }
 }