host P0.3b: durable flow log — survives restart (LIVE-VERIFIED)

The driver now persists each effect record to the blog store (string-keyed to dodge the keyword/
persist top-level split), and host/blog-load-flowlog! rebuilds the in-memory log on boot (wired into
serve.sh after load-edges!). So /flows survives a restart — closing the P0.3 gap.

LIVE PROOF: published a post on blog.rose-ash.com → /flows showed validate+notify → RESTARTED the
container (in-memory log lost) → /flows STILL showed them, reloaded from the durable store.
Round-trip also covered by a conformance test (persist → clear → reload → identical). blog 208/208,
full host conformance 599/599. Note: whole-list rewrite per effect — fine at P0 volume, cap/rotate later.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

lib/host/serve.sh

@@ -149,6 +149,11 @@ EPOCH=1
   echo "(epoch $EPOCH)"
   echo "(eval \"(host/blog-load-edges!)\")"
   EPOCH=$((EPOCH+1))
+  # P0.3b: rebuild the in-memory publish flow log from the durable store, so /flows
+  # survives a restart (the driver persists each effect record under "flowlog").
+  echo "(epoch $EPOCH)"
+  echo "(eval \"(host/blog-load-flowlog!)\")"
+  EPOCH=$((EPOCH+1))
   # Sessions on the DURABLE store, LAZILY: only a logged-in session (one that
   # writes a field) persists, so a login survives a restart while anonymous /
   # crawler traffic leaves no rows. host/session-init! bumps the per-boot epoch