Write get_baseline_predictions function

[lshandross]

Write function that generates predictions for one baseline model, one location:

get_baseline_predictions(target_ts, transformation, symmetrize, window_size, effective_horizons, quantile_levels, n_samples)

Fit single baseline model to existing target data, make future predictions (as samples from a probabilistic distribution) using that fit using stats::predict, then extract quantiles and/or samples. The extracted forecasts contain length (effective_horizons) * length (quantiles) numbers, but are stored as a nested list column in a 1x1 data frame

Inputs:

• target_ts a data frame of target data in a time series format (contains columns time_index, location, and observation)
• transformation a string specifying the transformation used to create the model. May be one of "sqrt" or "none"
• symmetrize boolean specifying whether or not to symmetrize the model
• window_size an integer specifying the window size used to create the model
• effective_horizons A numeric vector of prediction horizons relative to the last observed date in target ts
• quantile_levels numeric vector of quantile levels to output. Set to NA if quantile outputs not requested
• n_samples integer of samples to output (and to predict). Set to NA if sample outputs not requested, in which case 100,000 are generated by default