Stage G: SEND, and two workers interleave; decision compares work heat (FABRIC-3.8 §3.8, §3.13.7)

SEND ( paddr plen name-c name-u -- ok? ): general kernel-Hermes send to a
named VM, from = calling VM, one block max (refused, not truncated), every
VM. New SK_HERMES_MSG_TYPE_SEND (30).

Ruling C-4 (beta), Captain Bob: the switch decision compares work heat --
the heat each VM pulled in by its own per-beat charges -- not absolute
heat, so Hera's structural endowment no longer makes her hottest. Before
it, the two-worker test ran serial (AAAA...BBBB); after it, byte-identical
on amd64/aarch64/riscv64:
AAABBBBBBBBAAAAAAAABBBBBAAAAABBBBBAAAAABBBBBAAAAABBBBBAAAAABBBBBAAAAABBBBBAAAABB
(40 A, 40 B, 16 runs; ordinary words only, nothing yields). Acceptance:
relay 3, idle growth 0; dict_hash set changed by the new word
(word_count 533) and identical across ISAs.

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

include/starkernel/vm/kernel_hermes.h

@@ -752,6 +752,14 @@ int sk_hermes_reassemble(SkHermesMessage **chunks, int n_chunks,
  * Phase 4 entry. */
 #define SK_HERMES_MSG_TYPE_ELEVATE_REQUEST 8
 
+/* SK_HERMES_MSG_TYPE_SEND (FABRIC-3.8.md §3.8, Stage G, 2026-09-23) --
+ * a general message from the FORTH SEND word (repl.c): payload is FORTH
+ * text the recipient interprets, like CONSOLE_CMD, but from any VM to any
+ * VM. A fresh kernel-Hermes-space value, clear of messaging.4th's old type
+ * space and the negotiation types (20-24) listed below -- no FORTH-side
+ * type ever existed for it to reuse. */
+#define SK_HERMES_MSG_TYPE_SEND 30
+
 /* The five negotiation types above (20-24) were chosen clear of
  * messaging.4th's own live/reserved type space (PAUSE-EVENT=2,
  * RESUME-EVENT=3, KILL-EVENT=4, CONSOLE-CMD-EVENT=7, ELEVATE-REQUEST=8,