There were no tests for the encoders module.  encode_base64 worked
because it is the default and so got tested implicitly elsewhere, and
we use encode_7or8bit internally, so that worked, too.  I previously
fixed encode_noop, so this fix means that everythign in the encoders
module now works, hopefully correctly.  Also added an explicit test
for encode_base64.

The new _has_surrogates code was suggested by Serhiy Storchaka.  See
the issue for timings, but it is far faster than any other alternative,
and also removes the load time that we previously incurred from compiling
the complex regex this replaces.

On AIX, the C function mktime() alwaysd sets tm_wday, even on error. So tm_wday
cannot be used as a sentinel to detect an error, we can only check if the
result is (time_t)-1.

(test.regrtest and test.script_helper enable faulthandler module in
subprocesses).

in subprocesses.

time.strtime("%Y") returned "2345" when formatting year 12345.

time.strtime("%Y") returned "2345" when formatting year 12345.

the input string in longer than 2 gigabytes, and
ssl.SSLContext.load_cert_chain() raises a ValueError if the password is longer
than 2 gigabytes. The ssl module does not support partial write.

string in longer than 2 gigabytes, and ssl.SSLContext.load_cert_chain() raises
a ValueError if the password is longer than 2 gigabytes. The ssl module does
not support partial write.

"import idlelib" should not install hooks on the warning modules, hooks should
only be installed when IDLE is started.

of socket.socket objects now truncate the input buffer to INT_MAX bytes on
Windows to avoid an integer overflow.

(sendall() still send the whole buffer.)