chore: re-write evaluation in preparation for tuning

[DeveloperPaul123]

Changes

• Made PSQT and evaluation types public
• Added traits to make eval more generic so we can re-use more in the tuner
• Various tweaks to some structs in the chess crate.
• Re-implemented evaluation using new traits and other helpers. PSQT tables now stored as phased scores with piece values baked it.

Elo   | -0.25 +- 2.59 (95%)
SPRT  | 8.0+0.08s Threads=1 Hash=16MB
LLR   | 2.96 (-2.94, 2.94) [-5.00, 0.00]
Games | N: 33408 W: 9539 L: 9563 D: 14306
Penta | [1049, 3857, 6953, 3759, 1086]
https://pyronomy.pythonanywhere.com/test/520/

bench: 1583604