Split pipelines as smallest units for a given task

[Bisaloo]

Pipelines should focus on very specific tasks. But right now, the majority of the transmissibility pipeline focuses on descriptive epidemiology (case count summary, epicurves, etc.) rather than the Rt estimation.
This generates duplication across pipelines (e.g., both the transmissibility and severity pipelines would have the early linelist -> incidence step) and might actually be more bothersome for non-technical users who start directly with incidence data.

After an internal discussion, we decided to split the transmissibility pipeline into:

• Data preparation and descriptive epidemiology:
  • start from messy linelist data
  • clean data
  • aggregate data via incidence2
  • plot epicurves
  • plot cases on a map (Investigate ggautomap for descriptive statistics pipeline #94)
• Transmissibility pipeline:
  • start from incidence data
  • generate global Rt estimate
  • generate regional Rt estimates