let the python handle the loop while reading a byte array

[mdumandag]

we were looping over the result to convert the byte array into a list. This was causing a huge latency problem when the result we iterate over is huge. With this change, the conversion is delegated to python itself.

By the way, read_byte_array method returns list of byte sized integers, not a byte array. I think we can just return the result without converting it to a list since bytearray is special type in python. If we do that, we can also change the read_utf method to something like this as we did in the Node.js client

def read_utf(self):
  result = self.read_byte_array()
  if result is None:
    return None
  return result.decode("utf-8")

This read_utf change solves the issue described in #149. The duration of the string read test described in the issue goes from 35 seconds to 100-200 milliseconds in my laptop.