lint: Fix incorrect line length calculations in Python 2.

Previously, we would incorrectly be counting bytes in Python 2, which
meant lines with unicode characters in them appeared to the linter to
be far longer than they actually were.

tools/lint-all

@@ -157,7 +157,11 @@ def build_custom_checkers(by_lang):
                 except Exception:
                     print("Exception with %s at %s line %s" % (rule['pattern'], fn, i+1))
                     traceback.print_exc()
-            if (max_length is not None and len(line) > max_length and
+            if isinstance(line, bytes):
+                line_length = len(line.decode("utf-8"))
+            else:
+                line_length = len(line)
+            if (max_length is not None and line_length > max_length and
                 '# type' not in line and 'test' not in fn and 'example' not in fn and
                 not re.match("\[[a-z0-9_-]*\]: http.*", line) and
                     "#ignorelongline" not in line and 'migrations' not in fn):