Make the parallel operations more continuous

Instead of batching the list items, change the batch size to a
parallelism limit and run up to that many fibers. When the processing
of one list item finishes, another will then start immediately after,
rather than when the whole batch is finished.

These changes also make the fibers-map and fibers-for-each operations
work with vectors as well as lists.

tests/parallelism.scm

@@ -3,6 +3,7 @@
              (unit-test)
              (knots parallelism))
 
+;; Test fibers-map
 (run-fibers-for-tests
  (lambda ()
    (assert-equal
@@ -12,4 +13,34 @@
                 (* 2 i))
               (iota 34))))))
 
+;; Test fibers-batch-map with a large batch size
+(run-fibers-for-tests
+ (lambda ()
+   (assert-equal
+    1122
+    (apply + (fibers-batch-map
+              (lambda (i)
+                (* 2 i))
+              100
+              (iota 34))))))
+
+;; Test fibers-map with vectors
+(run-fibers-for-tests
+ (lambda ()
+   (assert-equal
+    1122
+    (apply + (vector->list
+              (fibers-map
+               (lambda (i)
+                 (* 2 i))
+               (list->vector (iota 34))))))))
+
+;; Test fibers-for-each
+(run-fibers-for-tests
+ (lambda ()
+   (fibers-for-each
+    (lambda (i)
+      (* 2 i))
+    (iota 34))))
+
 (display "parallelism test finished successfully\n")