Series wrapper for timeseries data with pipe-friendly processing capabilities.

Terminal operations return computed values:

series
|> aggregate_sum("count")
|> then(&IO.inspect(&1, label: "Total Count"))

timeline_data = series |> format_timeline("count")
categories = series |> format_category("status")

Transformation operations return a new %Trifle.Stats.Series{}:

new_series = series
|> transform_expression(["sum", "count"], "a / b", "avg")
|> transform_expression(["success", "total"], "(a / b) * 100", "success_rate")

result = series
|> transform_expression(["sum", "count"], "a / b", "avg")
|> transform_expression(["success", "total"], "(a / b) * 100", "rate")
|> aggregate_max("rate")