Syntax / macros for slices in numpy/pandas/etc.

[asemic-horizon]

I'm doing a Kaggle contest this weekend and noting a couple of patterns in typical EDA Python work that are awkward in Hy.

1. Multidimensional slicing like X[i,j], X[a:b,j], X[i,:].
2. Conditional slices like X[X.type == "N"]
3. Chained slices like X[X.type == "N"]["value"]

The latter, for example, looks like

 (get (get X (= (get X "type") "N")) "value"))

Can we consider this a feature request? It takes learning some numpy to figure out what are the explicit functions behind the syntax sugar, so it may not get done in any foreseeable future -- but it's a common Python use case and thus a major selling point for Hy.

[Kodiologist]

Duplicate of #541.