strategy.backtest without `to` argument should infer `start_time` and `end_time`

[working4coins]

Hi,

Running RSI example with

def main():
    # Define exchange as KeylessExchange
    exchange = blankly.KeylessExchange(
        price_reader=blankly.data.data_reader.PriceReader("./XBTUSDT_1D.csv", "BTC-USD")
    )

    # Use our strategy helper on Binance
    strategy = blankly.Strategy(exchange)

    # Run the price event function every time we check for a new price - by default that is 15 seconds
    strategy.add_price_event(price_event, symbol='BTC-USD', resolution='1d', init=init)

    #strategy.start()
    strategy.backtest(initial_values={'USD': 10000})

if __name__ == "__main__":
    main()

ie when no to = "1y" argument is passed

raises

backtest-run_trading-1  |   File "/usr/local/lib/python3.11/site-packages/blankly/exchanges/interfaces/paper_trade/backtest_controller.py", line 384, in sync_prices
backtest-run_trading-1  |     end_time = self.__user_added_times[i][self.PriceIdentifiers.epoch_stop] - resolution
backtest-run_trading-1  |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~
backtest-run_trading-1  | TypeError: unsupported operand type(s) for -: 'NoneType' and 'int'

Ideally in such a case backtest should use as much data as it can and define start_time and end_time automatically from data

If there is several PriceReader start_time should be max of start_time accross PriceReaders.
and end_time should be min of end_time to consider that a backtest can be done only for overlapping data.

Kind regards