fix(v4.0.0): step 6e review round 2 -- only the message looked at is taken, moved or dropped; the kernel's port has no queue in a fabric; an answer nobody awaits is not a loss

- Engine: a wire keeps whether the message at its mark is the one last
  looked at; a node's take, move and drop of any other answer 3 and write
  nothing (MESH.md 7d.3).  A wire cut and made again between a look and a
  take no longer has a NACK done as text, or 256 cells written where one
  was looked for.
- Engine: v4_fabric_wire_kernel wires a kernel to a port with no queues;
  a message for it is refused at once, error 18, on a mesh as on the
  products.
- The result of a drop is looked at; a take that does not take has the
  message dropped, or gone past: nothing goes round.
- An answer nobody waits for is let go uncounted; (LOST) counts what should
  have been delivered or acted on and was not.  The flood's account is
  exact.
- Error 18 from under a word that was printing empties the data stack, as
  a fault does: that word's own values are not left there.
- Tests: the take and drop windows cut at every step, with and without
  another message put; a refusal owed to a stuck node whose wire is full.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

v4/tests/test_host_unit.c

@@ -143,7 +143,7 @@ static int host_born(void *self)
     v4_blocks_init(&node_blocks[k], BUF0_W);
     if (k != 0) {                                    /* Hera's kernel is wired where she is made */
         node_kernel[k].take = node_kernel_take; node_kernel[k].give = 0; node_kernel[k].self = &pool[k]; node_kernel[k].pending = 0;
-        if (!v4_fabric_wire_device(&f, (unsigned)place, 0, &node_kernel[k])) return -1;
+        if (!v4_fabric_wire_kernel(&f, (unsigned)place, 0, &node_kernel[k])) return -1;   /* it takes words, and no messages */
     }
     return place;
 }
@@ -405,7 +405,7 @@ int main(void)
     hera_place = (unsigned)place;
     for (i = 0; i < 4096 && pool[0].n.mem[i] == 0; i++) { }
     CHECK(i == 4096 && pool[0].n.p == PORT + (v4_cell)V4_PORT_ANY, "with nothing in it, listening at its ports");
-    CHECK(v4_fabric_wire_device(&f, hera_place, 0, &kernel) && v4_fabric_wire_device(&f, hera_place, 1, &console), "the kernel is on its port 0 and the console on its port 1");
+    CHECK(v4_fabric_wire_kernel(&f, hera_place, 0, &kernel) && v4_fabric_wire_device(&f, hera_place, 1, &console), "the kernel is on its port 0 and the console on its port 1");
     for (i = 0; i < 400000 && v4_fabric_step(&f) != 0; i++) { }
     CHECK(given_cells == f18_count && pool[0].n.doing && pool[0].n.do_op == 8 && !pool[0].n.stopped,
           "it takes in the nucleus through its port, a word at a time, and sleeps until a message comes: %u of %u cells", given_cells, f18_count);
@@ -490,8 +490,7 @@ int main(void)
         CHECK(strcmp(tell(11, ": X S\" 66 EMIT\" 14 SEND ; X"), "B") == 0 && printed_from == 14, "and to the one across from it, which it is not wired to: \"%s\"", printed);
         CHECK(clock_of(10) > hera, "that went by a node in between");
     }
-    CHECK(strcmp(tell(10, "(LOST) @ ."), "0 ") == 0 && strcmp(tell(11, "(LOST) @ ."), "2 ") == 0 && strcmp(tell(14, "(LOST) @ ."), "0 ") == 0,
-          "nothing was let go but the two answers that came back to node 11, which did not wait for them: \"%s\"", printed);
+    CHECK(strcmp(tell(10, "(LOST) @ ."), "0 ") == 0 && strcmp(tell(11, "(LOST) @ ."), "0 ") == 0 && strcmp(tell(14, "(LOST) @ ."), "0 ") == 0, "nothing was let go");
 
     /* ---- they share the store (acceptance 3) ---- */
     step_limit = 30000000ul;
@@ -515,7 +514,7 @@ int main(void)
     CHECK(told(10, ": GO S\" W\" 12 SEND W ; GO"), "Hera sets node 12 writing and writes her own meanwhile: \"%s\"", printed);
     for (i = 0; i < 10 && common_dev[(12 + i) * V4_BLOCK_BYTES] == 'a' && common_dev[(22 + i) * V4_BLOCK_BYTES] == 'b'; i++) { }
     CHECK(i == 10, "all twenty blocks reached the store: the first that did not is %u", i);
-    CHECK(strcmp(tell(10, "(LOST) @ ."), "1 ") == 0 && strcmp(tell(12, "(LOST) @ ."), "0 ") == 0, "and no message was let go but node 12's answer to Hera, who did not wait for it: \"%s\"", printed);
+    CHECK(strcmp(tell(10, "(LOST) @ ."), "0 ") == 0 && strcmp(tell(12, "(LOST) @ ."), "0 ") == 0, "and no message was let go");
     CHECK(waiting(10) && waiting(11) && waiting(12) && waiting(13) && waiting(14), "all five are waiting at their ports again");
 
     /* ---- Hera kills a node by number, and tells the others it is gone (MESH.md 7b, 7d) ----
@@ -548,6 +547,8 @@ int main(void)
     CHECK(waiting(12) && waiting(11) && waiting(10), "every node that is not stuck is asleep again");
     CHECK(strcmp(tell(12, "12 100 * ."), "1200 ") == 0 && strcmp(tell(11, "11 100 * ."), "1100 ") == 0 && strcmp(tell(10, "1 2 + ."), "3 ") == 0, "and everything else keeps running: \"%s\"", printed);
     CHECK(strstr(tell(12, "5 3 PORT!"), "No one on that port") != NULL && waiting(12), "a write to the port node 14 was on is an error at once, not a wait: \"%s\"", printed);
+    CHECK(strstr(tell(12, ": K0 S\" 1 DROP\" 0 0 SEND-ON ; K0"), "No one on that port") != NULL && strcmp(tell(12, "2100 BLOCK C@ ."), "65 ") == 0,
+          "a message sent by the port node 12's kernel is on is refused at once -- the kernel takes no messages -- and its blocks are asked for as before: \"%s\"", printed);
     {
         v4_cell refused = node_numbered(12)->mem[REFUSED], r11 = node_numbered(11)->mem[REFUSED];
         CHECK(tell(12, ": AGAIN14 S\" 65 EMIT\" 14 SEND ; AGAIN14")[0] == 0 && node_numbered(12)->mem[REFUSED] == refused + 1 && waiting(12),