Research, creation and improvement of main search algorithm

[Jlisowskyy]

TODOS:

• There is probably some bug inside the search or the evaluation which significantly affects the overall performance, this should be investigated deeply
• There is some small problem with performance, in details we lost some speed, when it should not happen
• Investigate null move inside the TT
• Apply adjusting mate values distance from TT reads
• Implemented templated search function to simplify and squash previously used 4 search sub functions into 2 shorter ones.
• Added simple implementation of Internal Iterative deepening.
• Changed previous move picking function to perform in 'stable' manner
• Fixed small bug inside mate distance counting
• Added checking promotions inside the quiesce search
• Add SSE inside the quiesce
• Test and derive good enough parameters for Aspiration windows
• Add delta prunning to the quiesce
• Add prunning to the main search
• Add extensions
• Provide robust LMR reduction function #66
• Improve TT performance

Notes on the search algorithm:

• Instead of averaging all evals inside the aspiration window framework we could do something like resets on no pv hit
• Temporarly disabled Aspiration Windows due to search instability

[Jlisowskyy]

Found commit that slows down the search accidentally: 9a8929e

[Jlisowskyy]

Consider removing steady clock from search and optimizing streams?

[Jlisowskyy]

Found commit that slows down the search accidentally: 9a8929e

It showed out that everything is fine, the slow down was caused by some previous fixes

[Jlisowskyy]

There is some strange behavior around mate founding