Uniformize UTF-8 naming (#1115)

* Uniformize UTF-8 naming

Replace `utf8` -> `utf-8` everywhere.
It should have no impact, `utf8` is an alias of `utf-8` [1].

[1] https://github.com/python/cpython/blob/ee03bad25e83b00ba5fc2a0265b48c6286e6b3f7/Lib/encodings/aliases.py#L534

* Always specify the encoding

Let's be explicit over implicit. And prevent future warnings from PEP-597 [1].

[1] https://www.python.org/dev/peps/pep-0597/#using-the-default-encoding-is-a-common-mistake

* Update `UTF8` constant (`utf-8` -> `utf_8`)

* Remove default argument from `str.encode()` and `bytes.decode()`

* Clean-up

tests/utils/__init__.py

@@ -281,14 +281,14 @@ def http(
         output = stdout.read()
         devnull_output = devnull.read()
         try:
-            output = output.decode('utf8')
+            output = output.decode()
         except UnicodeDecodeError:
             r = BytesCLIResponse(output)
         else:
             r = StrCLIResponse(output)
 
         try:
-            devnull_output = devnull_output.decode('utf8')
+            devnull_output = devnull_output.decode()
         except Exception:
             pass