feat: core.async Phase 3 — channel transducers + dropping/sliding buffers

(chan n xform) applies a transducer on the put side: a jolt transducer is a
directly-callable closure, composed over a reducing fn whose step gives each
output value into the channel (honoring its buffer kind). One put may yield zero
or more values; a reduced result (e.g. from take) closes the channel; close!
runs the transducer completion arity to flush stateful remainders. Works with
map/filter/mapcat/take/comp/etc.

Buffers: (buffer n) fixed, (dropping-buffer n) drops new values when full,
(sliding-buffer n) drops the oldest. Implemented via a non-blocking give —
(ev/select [ch v] closed-chan) detects a full buffer without parking.

harness: run-spec flushes per suite. spec: core.async/channel-transducers (5),
core.async/buffers (3). jpm test green.

Note: distinct/dedupe/partition-all/partition-by still lack a 0-coll transducer
arity in core (separate gap), so they can't yet be used as channel xforms.

test/support/harness.janet

@@ -54,6 +54,7 @@
           (= (r 1) true) (++ pass)
           (array/push fails [label (string "want " expected ", got " (show actual))])))))
   (printf "  %s: %d/%d" suite pass (length cases))
+  (flush)
   (each [l m] fails (printf "    FAIL [%s] %s" l m))
   (when (> (length fails) 0)
     (error (string suite ": " (length fails) " failing behavior(s)")))