review uses of vapply and refactor where duplicated

[njtierney]

e.g.,

https://github.com/greta-dev/greta/blob/master/R/simulate.R#L65

target_node_names <- vapply(target_nodes,
    member,
    "unique_name",
    FUN.VALUE = character(1)
  )
  object_node_names <- vapply(object$dag$node_list,
    member,
    "unique_name",
    FUN.VALUE = character(1)
  )

and

https://github.com/greta-dev/greta/blob/master/R/utils.R#L504

can be replaced with the new base::lengths

[njtierney]

In vapply, FUN.VALUE is sometimes “” and other times, character(1), is there a difference/benefit from doing this?

[goldingn]

Same result, but I think I switched to character(1) as it makes the intent clearer

[njtierney]

Some common vapply patterns:

• vapply(x, member, "unique_name") (9)
• vapply(x, identical, ...) (6)
• vapply(x, is.null, ...) (5)
• vapply(x, is.greta_array) (5)
• vapply(x, has_distribution, ...) (4)
• vapply(dot_nodes, is_scalar, logical(1)) (2)
• vapply(x, member, "discrete") (2)
• classes <- vapply(x, class, "") (1)
• vapply(x, member, "multivariate") (1)
• vapply(names, deparse, "")
• vapply(nodes, dag$tf_name, ...)