feat(raft): add membership proposals, snapshot checkpoints, and log compaction (B4+B5+B6)

- ProposeAddPeerAsync/ProposeRemovePeerAsync: single-change-at-a-time membership
  changes through RAFT consensus (Go ref: raft.go:961-1019)
- RaftLog.Compact: removes entries up to given index for log compaction
- CreateSnapshotCheckpointAsync: creates snapshot and compacts log in one operation
- DrainAndReplaySnapshotAsync: drains commit queue, installs snapshot, resets indices
- Pre-vote protocol skipped (Go NATS doesn't implement it either)
- 23 new tests in RaftMembershipAndSnapshotTests

src/NATS.Server/Raft/RaftLog.cs

@@ -7,6 +7,11 @@ public sealed class RaftLog
 
     public IReadOnlyList<RaftLogEntry> Entries => _entries;
 
+    /// <summary>
+    /// The base index after compaction. Entries before this index have been removed.
+    /// </summary>
+    public long BaseIndex => _baseIndex;
+
     public RaftLogEntry Append(int term, string command)
     {
         var entry = new RaftLogEntry(_baseIndex + _entries.Count + 1, term, command);
@@ -28,6 +33,21 @@ public sealed class RaftLog
         _baseIndex = snapshot.LastIncludedIndex;
     }
 
+    /// <summary>
+    /// Removes all log entries with index &lt;= upToIndex and advances the base index accordingly.
+    /// This is log compaction: entries covered by a snapshot are discarded.
+    /// Go reference: raft.go WAL compact / compactLog.
+    /// </summary>
+    public void Compact(long upToIndex)
+    {
+        var removeCount = _entries.Count(e => e.Index <= upToIndex);
+        if (removeCount > 0)
+        {
+            _entries.RemoveRange(0, removeCount);
+            _baseIndex = upToIndex;
+        }
+    }
+
     public async Task PersistAsync(string path, CancellationToken ct)
     {
         Directory.CreateDirectory(Path.GetDirectoryName(path)!);