Disable a (not-really-for-Cython) test that fails due to Unicode fixes in Py3.8.

tests/run/test_unicode.pyx

@@ -842,7 +842,8 @@ class UnicodeTest(CommonTest,
         self.assertEqual('h\u0130'.capitalize(), 'H\u0069\u0307')
         exp = '\u0399\u0308\u0300\u0069\u0307'
         self.assertEqual('\u1fd2\u0130'.capitalize(), exp)
-        self.assertEqual('finnish'.capitalize(), 'FInnish')
+        # Note: differs between Py<3.8 and later.
+        #self.assertEqual('finnish'.capitalize(), 'FInnish')
         self.assertEqual('A\u0345\u03a3'.capitalize(), 'A\u0345\u03c2')
 
     def test_title(self):

tests/run/test_unicode_string_tests.pxi

@@ -977,8 +977,9 @@ class CommonTest(BaseTest):
     def test_capitalize_nonascii(self):
         # check that titlecased chars are lowered correctly
         # \u1ffc is the titlecased char
-        self.checkequal('\u03a9\u0399\u1ff3\u1ff3\u1ff3',
-                        '\u1ff3\u1ff3\u1ffc\u1ffc', 'capitalize')
+        # Note: differs between Py<3.8 and later.
+        #self.checkequal('\u03a9\u0399\u1ff3\u1ff3\u1ff3',
+        #                '\u1ff3\u1ff3\u1ffc\u1ffc', 'capitalize')
         # check with cased non-letter chars
         self.checkequal('\u24c5\u24e8\u24e3\u24d7\u24de\u24dd',
                         '\u24c5\u24ce\u24c9\u24bd\u24c4\u24c3', 'capitalize')