Tests: switch to explicit u"unicode" literals

Drop `from __future__ import unicode_literals`;
it was there for Python 3.2 compatibility (which
Anymail doesn't support). Ensures tests use normal
strs in Python 2.x.

tests/test_postmark_backend.py

@@ -1,7 +1,5 @@
 # -*- coding: utf-8 -*-
 
-from __future__ import unicode_literals
-
 from base64 import b64encode
 from decimal import Decimal
 from email.mime.base import MIMEBase
@@ -172,15 +170,13 @@ class PostmarkBackendStandardEmailTests(PostmarkBackendMockAPITestCase):
         self.assertNotIn('ContentID', attachments[2])
 
     def test_unicode_attachment_correctly_decoded(self):
-        # Slight modification from the Django unicode docs:
-        # https://django.readthedocs.io/en/latest/ref/unicode.html#email
-        self.message.attach("Une pièce jointe.html", '<p>\u2019</p>', mimetype='text/html')
+        self.message.attach(u"Une pièce jointe.html", u'<p>\u2019</p>', mimetype='text/html')
         self.message.send()
         data = self.get_api_call_json()
         self.assertEqual(data['Attachments'], [{
-            'Name': 'Une pièce jointe.html',
+            'Name': u'Une pièce jointe.html',
             'ContentType': 'text/html',
-            'Content': b64encode('<p>\u2019</p>'.encode('utf-8')).decode('ascii')
+            'Content': b64encode(u'<p>\u2019</p>'.encode('utf-8')).decode('ascii')
         }])
 
     def test_embedded_images(self):