Fix functional API tests to be endian-agnostic

Fix the "backslashreplace" tests for the functional API to be
endian-agnostic.  The tests used to rely on `.encode("utf-16")`
producing the same data as found in the test file.  However, on big
endian platforms it would produce a big endian encoding, while the test
file is little endian.  To avoid the problem, explicitly specify
`utf-16-le` encoding.  Since this meant that the BOM is no longer
produced, explicitly include it in input.

Fixes python#312

importlib_resources/tests/test_functional.py

@@ -82,7 +82,7 @@ def test_read_text(self):
                 'utf-16.file',
                 errors='backslashreplace',
             ),
-            'Hello, UTF-16 world!\n'.encode('utf-16').decode(
+            '\ufeffHello, UTF-16 world!\n'.encode('utf-16-le').decode(
                 errors='backslashreplace',
             ),
         )
@@ -130,7 +130,7 @@ def test_open_text(self):
         ) as f:
             self.assertEqual(
                 f.read(),
-                'Hello, UTF-16 world!\n'.encode('utf-16').decode(
+                '\ufeffHello, UTF-16 world!\n'.encode('utf-16-le').decode(
                     errors='backslashreplace',
                 ),
             )