bench: Add tests for constructing various containers from iterator.

Both "bound" (like, length known) and "unbound" (length unknown) are tested.
All of list, tuple, bytes, bytesarray offer approximately the same
performance, with "unbound" case being 30 times slower.

tests/bench/from_iter-1-list_bound.py

+import bench
+
+def test(num):
+    for i in iter(range(num//10000)):
+        l = [0] * 1000
+        l2 = list(l)
+
+bench.run(test)

tests/bench/from_iter-2-list_unbound.py

+import bench
+
+def test(num):
+    for i in iter(range(num//10000)):
+        l = [0] * 1000
+        l2 = list(map(lambda x: x, l))
+
+bench.run(test)

tests/bench/from_iter-3-tuple_bound.py

+import bench
+
+def test(num):
+    for i in iter(range(num//10000)):
+        l = [0] * 1000
+        l2 = tuple(l)
+
+bench.run(test)

tests/bench/from_iter-4-tuple_unbound.py

+import bench
+
+def test(num):
+    for i in iter(range(num//10000)):
+        l = [0] * 1000
+        l2 = tuple(map(lambda x: x, l))
+
+bench.run(test)

tests/bench/from_iter-5-bytes_bound.py

+import bench
+
+def test(num):
+    for i in iter(range(num//10000)):
+        l = [0] * 1000
+        l2 = bytes(l)
+
+bench.run(test)

tests/bench/from_iter-6-bytes_unbound.py

+import bench
+
+def test(num):
+    for i in iter(range(num//10000)):
+        l = [0] * 1000
+        l2 = bytes(map(lambda x: x, l))
+
+bench.run(test)

tests/bench/from_iter-7-bytearray_bound.py

+import bench
+
+def test(num):
+    for i in iter(range(num//10000)):
+        l = [0] * 1000
+        l2 = bytearray(l)
+
+bench.run(test)

tests/bench/from_iter-8-bytearray_unbound.py

+import bench
+
+def test(num):
+    for i in iter(range(num//10000)):
+        l = [0] * 1000
+        l2 = bytearray(map(lambda x: x, l))
+
+bench.run(test)