fix(s3): open an object's metadata and body stream together for CloudFront

The streaming change from upstream (#5278) serves S3 origins through
S3Service.openObject, which upstream added in a larger disk-streaming rework
(#4691) not taken here. This adds the same method on top of the existing
openObjectStream.

src/main/java/io/github/hectorvent/floci/services/s3/S3Service.java

@@ -1332,6 +1332,23 @@ public class S3Service implements Resettable, ResourceProvider {
         }
     }
 
+    /** An object's metadata together with an open stream of its body; the caller closes it. */
+    public record ObjectRead(S3Object object, InputStream body) implements java.io.Closeable {
+        @Override
+        public void close() throws IOException {
+            body.close();
+        }
+    }
+
+    /**
+     * Like {@link #getObject}, but the body is an open stream instead of a byte array, so a large
+     * object is served without being read into the heap.
+     */
+    public ObjectRead openObject(String bucketName, String key, String versionId) {
+        S3Object object = getObjectMetadata(bucketName, key, versionId);
+        return new ObjectRead(object, openObjectStream(bucketName, key, versionId));
+    }
+
     public InputStream openObjectStream(String bucketName, String key, String versionId) {
         getObjectMetadata(bucketName, key, versionId);
         String bucketOwnerAccount = resolveBucketEntry(bucketName)