Port XLISP-STAT super-smoother

XLISP-STAT had an [implementation](https://github.com/Lisp-Stat/xls-archive/tree/master/statistics/smoothers/supersmoother) of [Friedman's Super Smoother](http://fmwww.bc.edu/RePEc/bocode/s/supsmooth_doc.pdf) that would be nice to have in Lisp-Stat.

This is a good first issue for someone wishing to learn [CLOS](https://en.wikipedia.org/wiki/Common_Lisp_Object_System). The XLISP-Stat version has a [Flavors](https://en.wikipedia.org/wiki/Flavors_(programming_language)) based implementation. The conversion is relatively straightforward and provides an opportunity to both learn some of the OO history in Lisp and CLOS.

Start with the linked XLISP-STAT super-smoother implementation and the Friedman Super Smoother paper to understand the existing algorithm. Then review the repository's statistical smoothing functions and the Common Lisp Object System guidance in the issue. Done means a CLOS-based super-smoother is available in Lisp-Stat with behavior matching the referenced implementation.

Written by the indexing model from the issue text.