add support for parsing implicit DER tags

[tomato42]

While not needed for ECDSA/ECDH PKCS#8 files, it looks like it will be necessary for the PQC algorithms, so provide that too, so we can do some code reuse.

[rjrelyea]

I have a suggestion, which you can add before merge or ignore.

[rjrelyea]

We probably should have a test for length > 128 to make sure the llen encoding/decoding is happening correctly

[tomato42]

this is reusing common code for encoding and decoding length, so I'm quite sure that it will handle it correctly

we have test coverage for it here:

python-ecdsa/src/ecdsa/test_der.py

Lines 90 to 125 in 55d2b56

	class TestReadLength(unittest.TestCase):
	# DER requires the lengths between 0 and 127 to be encoded using the short
	# form and lengths above that encoded with minimal number of bytes
	# necessary
	def test_zero_length(self):
	self.assertEqual((0, 1), read_length(b"\x00"))
	def test_two_byte_zero_length(self):
	with self.assertRaises(UnexpectedDER):
	read_length(b"\x81\x00")
	def test_two_byte_small_length(self):
	with self.assertRaises(UnexpectedDER):
	read_length(b"\x81\x7f")
	def test_long_form_with_zero_length(self):
	with self.assertRaises(UnexpectedDER):
	read_length(b"\x80")
	def test_smallest_two_byte_length(self):
	self.assertEqual((128, 2), read_length(b"\x81\x80"))
	def test_zero_padded_length(self):
	with self.assertRaises(UnexpectedDER):
	read_length(b"\x82\x00\x80")
	def test_two_three_byte_length(self):
	self.assertEqual((256, 3), read_length(b"\x82\x01\x00"))
	def test_empty_string(self):
	with self.assertRaises(UnexpectedDER):
	read_length(b"")
	def test_length_overflow(self):
	with self.assertRaises(UnexpectedDER):
	read_length(b"\x83\x01\x00")