Small eval randomization for drawn terminal nodes (AndyGrant#135)

Idea from Stockfish

The small randomization, using the thread's node count, ensures that when different potential 3-fold lines are available, the search keeps exploring different lines. This helps to discover lines where one side can escape the 3-fold with advantage. This is especially useful at longer TCs.

ELO   | 1.28 +- 2.10 (95%)
SPRT  | 12.0+0.12s Threads=1 Hash=8MB
LLR   | -0.43 (-2.94, 2.94) [0.00, 4.00]
Games | N: 49920 W: 11997 L: 11813 D: 26110
http://chess.grantnet.us/viewTest/4797/

ELO   | 3.63 +- 2.88 (95%)
SPRT  | 60.0+0.6s Threads=1 Hash=64MB
LLR   | 2.97 (-2.94, 2.94) [0.00, 5.00]
Games | N: 21050 W: 4092 L: 3872 D: 13086
http://chess.grantnet.us/viewTest/4799/

BENCH : 5,132,900

Add MultiPV information to the README

NO FUNCTIONAL CHANGE

BENCH : 8,790,188

Cleanup board.x evaluate.x search.x transposition.x

Final work for AndyGrant#84

ELO   | 3.22 +- 3.63 (95%)
SPRT  | 10.0+0.1s Threads=1 Hash=8MB
LLR   | 2.95 (-2.94, 2.94) [-3.00, 1.00]
Games | N: 14480 W: 3059 L: 2925 D: 8496
http://chess.grantnet.us/viewTest/3447/

NO FUNCTIONAL CHANGE

BENCH : 7,065,845

Tune all linear eval terms using a new tuner

Convert tuning into the proper non-search based evaluation as per the original texel tuning. Tune using mini-batch GD with a batch size of 8192 and psuedo random shuffling of the tuning data.

ELO   | 5.87 +- 4.09 (95%)
SPRT  | 10.0+0.1s Threads=1 Hash=8MB
LLR   | 2.97 (-2.94, 2.94) [0.00, 4.00]
Games | N: 13440 W: 3376 L: 3149 D: 6915
http://chess.grantnet.us/viewTest/3012/

ELO   | 2.80 +- 2.25 (95%)
SPRT  | 60.0+0.6s Threads=1 Hash=64MB
LLR   | 2.96 (-2.94, 2.94) [0.00, 4.00]
Games | N: 36330 W: 7343 L: 7050 D: 21937
http://chess.grantnet.us/viewTest/3013/

BENCH : 7,503,296

Small cleanup for board.c and attacks.c. Add Laldon to authors

ELO   | 2.67 +- 4.36 (95%)
SPRT  | 10.0+0.1s Threads=1 Hash=8MB
LLR   | 2.95 (-2.94, 2.94) [-5.00, 0.00]
Games | N: 10400 W: 2265 L: 2185 D: 5950

NO FUNCTIONAL CHANGE

BENCH : 6,350,265

Remove useless call to hashfull()

NO FUNCTIONAL CHANGE

BENCH : 5,645,233

Refine safe definition, retune using new definition. Ethereal 10.00.

SPRT @ 5.0+0.05s Threads=1 Hash=8MB
LLR: 2.21 (-2.20, 2.20) [0.00, 5.00]
Games: 3550 W: 879 L: 761 D: 1910

SPRT @ 20.0+0.2s Threads=1 Hash=32MB
LLR: 2.20 (-2.20, 2.20) [0.00, 5.00]
Games: 2450 W: 518 L: 413 D: 1519

BENCH : 6,071,849

Added support for Syzygy through Fathom (Jon Dart's updated Fork)

I do not currently have an SSD, so testing was limited to a standard HD...

Testing with ProbeDepth=8

Ethereal 9.64: 4083 - 3751 - 11870  [0.508] 19704 @ 20+.2s and 8MB Hash

I've not yet seen any crashes running with TB. If you do, please inform me.

My hope is that the elo gain is much larger with the proper hardware.

Clean up some pruning in the move loop around MATED scores

This is really a bugfix, as it is possible (although rare) to return
a false MATED score when the first move searched is a MATED move,
we would trigger a research since this only happens in non PV nodes.
Thus, I do not feel the need to test this at LTC.

SPRT @ 5.0+0.05s Threads=1 Hash=2MB
LLR: 3.00 (-2.94, 2.94) [0.00, 5.00]
Games: 25675 W: 6956 L: 6670 D: 12049

BENCH : 9,484,343

Make the Texel Tuner use a tenth of the memory and about 7 times faster

NO FUNCTIONAL CHANGE

BENCH : 10,439,396