Added QAT v1.7 support

[dgarske]

• Added support for new usdm memory driver.
• Added support for detecting QAT API version and features.
• Added QAT_ENABLE_RNG option.
• Added QAT RSA Key Generation support.
• Added QAT SHA3 support.
• Improvements to QAT hashing support.
• Updates to README.md.

Intel QuickAssist 8950 (2.5-4x improvement)

Key Generation Benchmarks:

RSA   1024 key gen   SW 119.883 ops/sec
RSA   2048 key gen   SW 18.738 ops/sec
RSA   1024 key gen   HW 284.242 ops/sec
RSA   2048 key gen   HW 72.734 ops/sec

Intel QuickAssist 8970 (7.8-9x improvement)

Key Generation Benchmarks:

RSA   1024 key gen   SW 69.225 ops/sec
RSA   2048 key gen   SW 16.085 ops/sec
RSA   1024 key gen   HW 540.348 ops/sec
RSA   2048 key gen   HW 153.033 ops/sec

SHA-3 Benchmarks:

SHA3-256        SW   115 MB took 1.034 seconds,  111.256 MB/s Cycles per byte =  29.08
SHA3-256        SW   115 MB took 1.034 seconds,  111.251 MB/s Cycles per byte =  29.08
SHA3-256        SW   115 MB took 1.034 seconds,  111.191 MB/s Cycles per byte =  29.10
SHA3-256        SW   115 MB took 1.037 seconds,  110.850 MB/s Cycles per byte =  29.18
SHA3-256        SW   115 MB took 1.038 seconds,  110.822 MB/s Cycles per byte =  29.19
SHA3-256        SW   115 MB took 1.038 seconds,  110.802 MB/s Cycles per byte =  29.20
SHA3-256        SW   115 MB took 1.038 seconds,  110.795 MB/s Cycles per byte =  29.20
SHA3-256        SW   115 MB took 1.039 seconds,  110.666 MB/s Cycles per byte =  29.23
SHA3-256    SW  887.633 MB/s
SHA3-256        HW   315 MB took 1.002 seconds,  314.250 MB/s Cycles per byte =  10.29
SHA3-256        HW   315 MB took 1.004 seconds,  313.771 MB/s Cycles per byte =  10.31
SHA3-256        HW   310 MB took 1.004 seconds,  308.721 MB/s Cycles per byte =  10.48
SHA3-256        HW   320 MB took 1.005 seconds,  318.559 MB/s Cycles per byte =  10.16
SHA3-256        HW   340 MB took 1.005 seconds,  338.223 MB/s Cycles per byte =   9.57
SHA3-256        HW   335 MB took 1.006 seconds,  332.847 MB/s Cycles per byte =   9.72
SHA3-256        HW   320 MB took 1.007 seconds,  317.698 MB/s Cycles per byte =  10.18
SHA3-256        HW   315 MB took 1.008 seconds,  312.636 MB/s Cycles per byte =  10.35
SHA3-256    HW 2556.704 MB/s