define encoding when communicating with the gpg-agent process

This fixes:
  $ vault_from_gpg_agent.py --clear
  Traceback (most recent call last):
    File "/usr/local/bin/vault_from_gpg_agent.py", line 91, in <module>
      main()
    File "/usr/local/bin/vault_from_gpg_agent.py", line 85, in main
      clear_passphrase(gpg_agent, cache_id)
    File "/usr/local/bin/vault_from_gpg_agent.py", line 61, in clear_passphrase
      stdout = gpg_agent.communicate("CLEAR_PASSPHRASE %s\n" % (cache_id,))[0]
    File "/usr/lib/python3.7/subprocess.py", line 939, in communicate
      stdout, stderr = self._communicate(input, endtime, timeout)
    File "/usr/lib/python3.7/subprocess.py", line 1666, in _communicate
      input_view = memoryview(self._input)
  TypeError: memoryview: a bytes-like object is required, not 'str'

I set the encoding='utf8' when doing the `Popen`. As long as only ascii
characters get passed back and forth we should be safe.

Author: tpo

vault_from_gpg_agent.py

@@ -67,7 +67,7 @@ def main():
     parser.add_argument("--clear", default=False, action="store_true",
                         help="Clear password from GPG agent")
     gpg_agent = subprocess.Popen(["gpg-connect-agent"], stdin=subprocess.PIPE,
-                                 stdout=subprocess.PIPE)
+                                 stdout=subprocess.PIPE, encoding='utf8')
     args = parser.parse_args()
     my_path = os.path.realpath(sys.argv[0])
     # Per the source, cache-id is limited to 50 bytes, so we hash our