Request: once-and-for-all definitions of free-vars, free-in, not-free-in

I'll try to submit a patch in the coming weeks, but I'm going to leave this here for now in case someone knows how to do it more quickly than I do.

It seems like with the addition of binding-forms, we should be able to add a few more binding related (meta)functions "once-and-for-all", like was done for substitute and alpha-equivalent?.

In particular:

; Returns the free variables of any_e
free-vars : any_e -> (any_x ...)

; Returns #t when any_x is free in any_e, and #f otherwise.
free-in : any_e any_x -> #t or #f

; Returns #t when any_x is not free in any_e, and #f otherwise.
not-free-in : any_e any_x -> #t or #f

; Returns any_e with each any_xi replaced by any_ei. Substitution should be performed in order, so any any_e{i-1} can have any_xi.
; e.g.s
; (substitute* x (x y) (1 2)) -> 1
; (substitute* x (x y) (y 1)) -> 1
; (substitute* (x y) (x y) (1 2)) -> (1 2)
substitute* : any_e (any_xi ..._0) (any_ei ..._ 0) -> any_e

Review the existing binding-forms support and the implementations of substitute and alpha-equivalent? first. Compare the proposed free-vars, free-in, not-free-in, and substitute* interfaces with the examples in the issue, then confirm that the intended semantics and completion criteria are agreed before starting implementation.