scca_compute_tree: Build recursively a scca tree
In UtrechtUniversity/SCCA: Spectral Clustering Correspondence Analysis in R

See scca_compute for description

scca_compute_tree(
  labels,
  m,
  child,
  depth,
  max_depth,
  n_node,
  iter.max = 10,
  nstart = 50,
  max_eigenvalues = max_eigenvalues,
  decomp = "svd",
  heuristic = eigengap_heuristic
)

`labels`	The labels (character vector) defining the (sub-)set of data set m to be analyzed
`m`	A matrix representing a bi-partite network. The matrix must have row names and column names.
`child`	The child number of this node within its siblings (= nodes with same parent)
`depth`	The depth (integer) of this node in the tree.
`max_depth`	The maximum allowed depth of the analysis proces. If Inf (default) the analysis goes on untill a stop condition has been met.
`n_node`	Number of the node in the tree. This is a depth-first, pre-order numbering, starting with 1 at the top node (a.k.a. root)
`iter.max`	The maximum number of iterations kmeans is allowed to make. Default is 10.
`nstart`	Number of random cluster sets kmeans may choose to start with. Default is 25.
`max_eigenvalues`	Restrict the number of computed eigenvalues to max_eigenvalues. The default is 25.
`decomp`	The decomposition function to use. Choices are svd (default) and svd
`heuristic`	The function to use for calculating the number of clusters. The default is eigengap_heuristic

Function scca_compute_tree calls itself recursively for every sub cluster found at a node until one of the stop conditions is met:

The heuristic finds only one prominent eigenvalue (k = 1)
The maximum depth in a branch has been reached. k will be set to 0
The number of observations in the subset is too small to perform svd(s). This will raise a warning and k will be set to -1