knn: Classify points using k-nearest neighbors.
In nick-ulle/RCompilerExamples: Examples for RLLVMCompile

This version is non-idiomatic to suit the compiler's quirks, and should not be called directly in the R interpreter.

1	.knn(distances, labels, n_class, k, n_train, n_test)

`distances`	numeric, distance matrix, where rows correspond to training points and columns correspond to test points
`labels`	integer, class labels starting at 1
`n_class`	integer, number of classes
`k`	integer, number of nearest neighbors to use
`n_train`	integer, number of training points
`n_test`	integer, number of test points