Generation doesn't work for variant "Snort"

[Belzedar94]

The generator doesn't crash, but it stays at 0 progress forever even at low depths.

Example of command used:

Losd variants.ini
setoption name Threads value 28
setoption name Hash value 2048
setoption name UCI_Variant value snort
isready
generate_training_data depth 2 count 100000000 random_multi_pv 4 random_multi_pv_diff 300 random_move_count 8 random_move_max_ply 20 write_min_ply 5 eval_limit 10000 set_recommended_uci_options data_format bin output_file_name snort_100Md2.bin

[Belzedar94]

For Gale (That inherits rules from Snort), data generation works but produces some weird output that crashes during training.

Example:

fen 1B2B2B/2B2B2/2P2P2/P2kP2P/B2N2PB/1K1B4/P2P2P1/1P2PN1N b Qk - 8 2580 move a1h5 score -80 ply 5 result -1 e fen 1B2B2B/2B2B2/2P2P2/P2kP2P/B2N2PB/1K1B4/P2P2P1/1P2PN1N w Qk - 8 2580 move b1b1 score 208 ply 6 result 1 e fen 1B2B2B/2B2B2/2P2P2/P2kP2P/B2N2PB/1K1B4/P2P1P1P/2P2PN1 b KQ - 70 17028 move a1h8 score -80 ply 7 result -1 e fen 1B2B2B/2B2B2/2P2P2/P2kP2P/B2N2PB/1K1B4/P2P1P1P/2P2PN1 w KQ - 70 17028 move a1h8 score 208 ply 8 result 1 e fen 1B2B2B/2B2B2/2P2P2/P2kP2P/B2N2PB/1K1B4/P2P1P1P/2P2PN1 b KQ - 70 17028 move a1g7 score -80 ply 9 result -1 e fen 1B2B2B/2B2B2/2P2P2/P2kP2P/B2N2PB/1K1B4/P2P1P1P/2P2PN1 w KQ - 6 24964 move a1h8 score 208 ply 10 result 1 e fen 1B2B2B/2B2B2/2P2P2/P2kP2P/B2N2PB/1K1B4/P1P1P1P1/P2P2PN b Qk - 12 17160 move a1a7 score -80 ply 11 result -1 e

[ianfab]

I just merged the recent fixes from Fairy-SF, would be interesting to know if the problems still persist.