[enhancement] dynamically allocate `maxcpus`

[jpdeleon]

https://github.com/enzoampil/fastquant/blob/e0c6a7e4b18c16075a42cfaa5f711a522e4c8f1e/fastquant/strategies.py#L528

There is no decrease in run time using built-in grid search in backtest compared to basic loop-based grid search using identical parameter grid. It may be good to increase maxcpus if input parameter in backtest is a list else maxcpus=1 (default).

See commit ffdd2fc322475266c10668bbb31e58cb1bd9a9cd in auto_grid_search.