[query-service] teach query service to read MTs and Ts created by Spa…

…rk (#10184)

* [query-service] teach query service to read MTs and Ts created by Spark

Hail-on-Spark uses HadoopFS which emulates directories by creating size-zero files with
the name `gs://bucket/dirname/`. Note: the object name literally ends in a slash. Such files
should not be included in `listStatus` (they should always be empty anyway). Unfortunately,
my fix in #9914 was wrong because `GoogleStorageFileStatus` removes
the trailing slash. This prevented the path from matching `path`, which always ends in a `/`.

* fix

hail/src/main/scala/is/hail/io/fs/GoogleStorageFS.scala

@@ -317,8 +317,9 @@ class GoogleStorageFS(serviceAccountKey: String) extends FS {
     val blobs = storage.list(bucket, BlobListOption.prefix(path), BlobListOption.currentDirectory())
 
     blobs.getValues.iterator.asScala
-      .map(b => GoogleStorageFileStatus(b))
-      .filter(fs => !(fs.isDirectory && fs.getPath == path))
+      .map(b => (b, GoogleStorageFileStatus(b)))
+      .filter { case (b, fs) => !(fs.isDirectory && b.getName == path) } // elide directory markers created by Hadoop
+      .map { case (b, fs) => fs }
       .toArray
   }