Add Norm move class

[mwaskom]

A simple Move operation (slightly stretching the functionality beyond "reduce over-plotting") that does a divisive normalization by an arbitrary function.:

so.Plot(penguins, "bill_length_mm").add(so.Bar(), so.Hist(), so.Norm())

The function and grouping are flexible, and it can be chained with other moves in interesting ways:

(
    so.Plot(tips, x="day", y="total_bill", color="sex")
    .add(so.Bar(), so.Hist(), [so.Norm("sum", by=["x"]), so.Stack()])
)

There is also some flexibility to the domain across where the normalizer is computed:

(
    so.Plot(gapminder, "year", "lifeExp", color="continent")
    .add(so.Line(), so.Agg(), so.Norm(where="x == 1977"))
)

[codecov]

Codecov Report

Merging #2827 (90d14ad) into master (fefd940) will increase coverage by 0.00%.
The diff coverage is 100.00%.

@@           Coverage Diff           @@
##           master    #2827   +/-   ##
=======================================
  Coverage   98.04%   98.04%           
=======================================
  Files          68       68           
  Lines       21991    22042   +51     
=======================================
+ Hits        21561    21612   +51     
  Misses        430      430           

Impacted Files	Coverage Δ
seaborn/_core/moves.py	100.00% <100.00%> (ø)
seaborn/_core/plot.py	94.58% <100.00%> (+0.04%)	⬆️
seaborn/objects.py	100.00% <100.00%> (ø)
seaborn/tests/_core/test_moves.py	100.00% <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 fefd940...90d14ad. Read the comment docs.