Use the enc_flags field to determine whether we should use explicit IV,
signature algorithms or SHA256 default PRF instead of hard coding which
versions support each requirement.

Not sure why I am getting these now and not before.

The version check for DTLS1_VERSION was redundant as
DTLS1_VERSION > TLS1_1_VERSION, however we do need to
check for DTLS1_BAD_VER for compatibility.

PR:2984

Revert is appropriate because binary compatibility is not an issue
in 1.1.

MD5 should use little endian order. Fortunately the only ciphersuite
affected is EXP-RC2-CBC-MD5 (TLS_RSA_EXPORT_WITH_RC2_CBC_40_MD5) which
is a rarely used export grade ciphersuite.
(cherry picked from commit f306b87d766e6ecf30824635c7c395b67cff9dbc)

Address CBC decrypt timing issues and reenable the AESNI+SHA1 stitch.
(cherry picked from commit 125093b59f3c2a2d33785b5563d929d0472f1721)

Kludge alert. This is arranged by passing padding length in unused
bits of SSL3_RECORD->type, so that orig_len can be reconstructed.
(cherry picked from commit 8bfd4c659f180a6ce34f21c0e62956b362067fba)

(cherry picked from commit 04e45b52ee3be81121359cc1198fd01e38096e9f)

(cherry picked from commit d5371324d978e4096bf99b9d0fe71b2cb65d9dc8)

RISCs are picky and alignment granted by compiler for md_state can be
insufficient for SHA512.
(cherry picked from commit 36260233e7e3396feed884d3f501283e0453c04f)

Break dependency on uint64_t. It's possible to declare bits as
unsigned int, because TLS packets are limited in size and 32-bit
value can't overflow.
(cherry picked from commit cab13fc8473856a43556d41d8dac5605f4ba1f91)

(cherry picked from commit 34ab3c8c711ff79c2b768f0b17e4b2a78fd1df5d)

We have to use EVP in FIPS mode so we can only partially mitigate
timing differences.

Make an extra call to EVP_DigestSignUpdate to hash additonal blocks
to cover any timing differences caused by removal of padding.
(cherry picked from commit b908e88ec15aa0a74805e3f2236fc4f83f2789c2)

(cherry picked from commit 014265eb02e26f35c8db58e2ccbf100b0b2f0072)