improved Alpaca data process time

[mikazlopes]

I performed some changes in the Alpaca download_data, clean_data, add_technical_indicators, and add_vix, which reduce the processing time by 80%. I tested the data sanity by running both data collecting and cleaning processes side by side and comparing them using this function:

        def compare_data(data1, data2):
            return data1.equals(data2)


        is_equal = compare_data(data, data_downloaded)
        print(f"The data is equal: {is_equal}")
        (data == data_downloaded).all().all()
        different_rows = data[data != data_downloaded].dropna(how='all')
        print(different_rows)
        len(data) == len(data_downloaded)

And it returned True. I also changed a line in the data_processor.py code to allow it to use my trainer code at https://github.com/mikazlopes/training-farm, which can cache the clean data to avoid repeating the operation over and over using the same dataset. I also avoided using df.copy as much as possible to reduce RAM usage and add to the processing time.

[zhumingpassional]

Good codes.

thanks for your work.