feat: implement batch_store endpoint (Issue #12)

- Add batch_store tool accepting 1-50 entries per call
- Single DB transaction for atomicity
- Returns individual IDs/status per entry
- Add batch_store_memories() to Database layer
- Add 6 test cases
- Backward compatible - existing store unchanged

Expected impact: 50-60% reduction in store API calls

src/db.rs

@@ -174,3 +174,36 @@ impl Database {
         Ok(row.get("count"))
     }
 }
+
+
+/// Result for a single batch entry
+#[derive(Debug, Clone)]
+pub struct BatchStoreResult {
+    pub id: String,
+    pub status: String,
+}
+
+impl Database {
+    /// Store multiple memories in a single transaction
+    pub async fn batch_store_memories(
+        &self,
+        agent_id: &str,
+        entries: Vec<(String, Value, Vec<f32>, Vec<String>)>,
+    ) -> Result<Vec<BatchStoreResult>> {
+        let mut client = self.pool.get().await?;
+        let transaction = client.transaction().await?;
+        let mut results = Vec::with_capacity(entries.len());
+
+        for (content, metadata, embedding, keywords) in entries {
+            let id = Uuid::new_v4();
+            let vector = Vector::from(embedding);
+            transaction.execute(
+                r#"INSERT INTO memories (id, agent_id, content, embedding, keywords, metadata) VALUES ($1, $2, $3, $4, $5, $6)"#,
+                &[&id, &agent_id, &content, &vector, &keywords, &metadata],
+            ).await?;
+            results.push(BatchStoreResult { id: id.to_string(), status: "stored".to_string() });
+        }
+        transaction.commit().await?;
+        Ok(results)
+    }
+}