perf: defer HNSW persistence during addMany() batch operations

addMany() now temporarily switches the HNSW index to deferred persist
mode during the batch loop. Previously, each add() triggered ~16-20
saveHNSWData calls for modified neighbors (each a read→gzip→atomic-write
cycle). For 450 items this was ~8,100 individual storage writes.

With deferred mode, dirty node IDs are collected in a Set during the
batch and flushed once at the end — deduplicating repeated neighbor
updates. A node modified by insert #3 and again by insert #47 is
written only once.

Also adds setPersistMode() to TypeAwareHNSWIndex, propagating mode
changes to all existing type-specific sub-indexes.

src/hnsw/typeAwareHNSWIndex.ts

@@ -163,6 +163,19 @@ export class TypeAwareHNSWIndex {
     return this.persistMode
   }
 
+  /**
+   * Switch persist mode at runtime. Propagates to all existing type-specific indexes.
+   * Used by addMany() to defer persistence during batch operations.
+   */
+  public setPersistMode(mode: 'immediate' | 'deferred'): void {
+    this.persistMode = mode
+    for (const index of this.indexes.values()) {
+      if (typeof (index as any).setPersistMode === 'function') {
+        (index as any).setPersistMode(mode)
+      }
+    }
+  }
+
   /**
    * Get or create HNSW index for a specific type (lazy initialization)
    *