Ensure clim window has same size as skill today and place forecast_prev update at end of extrapolation loop

[RubenImhoff]

This PR makes two minor changes that impact the climatological skill calculation and the extrapolation loop in the steps blending code when subtimesteps are used.

Changes:

• When the climatological skill window is saved, it first checks if the arrays have the same size as the skill that will be appended. Before, this resulted in an error when moving to the next day (skill clim window gets saved) when the new forecasts have a different number of ensemble members than the skill clim window. This has been solved now.
• The update of "forecast_prev", the extrapolation and noise cascade that are fed into the extraplation loop, took place too early in the loop. This caused that the extrapolation loop already uses the next forecast when subtimesteps are used. This PR fixes this issue, too.

[codecov[bot]]

Codecov Report

Merging #274 (b553bb2) into master (8936236) will not change coverage. The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master     #274   +/-   ##
=======================================
  Coverage   82.32%   82.32%           
=======================================
  Files         158      158           
  Lines       12130    12130           
=======================================
  Hits         9986     9986           
  Misses       2144     2144           

Flag	Coverage Δ
unit_tests	82.32% <100.00%> (ø)

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
pysteps/blending/clim.py	100.00% <ø> (ø)
pysteps/blending/steps.py	83.73% <100.00%> (ø)

---

Continue to review full report at Codecov.

Legend - Click here to learn more Δ = absolute <relative> (impact), ø = not affected, ? = missing data Powered by Codecov. Last update 8936236...b553bb2. Read the comment docs.