So far we've overridden almost all string methods, that bstr/ustr
inherited from bytes and unicode. However 2 of the methods remained
intact until now: unicode.encode() and bytes.decode(). Let's override
them too for completeness:

- we want ustr.encode() to follow signature of unicode.encode and for ustr.encode('utf-8') to return bstr.
- for consistency we also want ustr.encode() to return the same type
  irregardless of which encoding/errors pair is used in the arguments.
- => ustr.encode() always returns bstr.
- we want bstr.decode() to follow signature of bytes.decode and for bstr.decode('utf-8') to return ustr.
- for consistency we also want bstr.decode() to return the same type
  irregardless of which encoding/errors pair is used in the arguments.
- -> bstr.decode() always returns ustr.

So  ustr.encode() -> bstr  and  bstr.decode() -> ustr.

Let's implement this carrying out encoding/decoding process internally
similarly to regular bytes and unicode and wrapping the result into
corresponding pygolang type at the end.