Lay the prompt out so the endpoint can cache most of it

Prompt caching bills on a shared prefix: the endpoint reuses the request up
to the first byte that differs from last time and no further. The live
world-state block sat third from the top of the system message, so every turn
re-priced the instructions, the plot essentials and the whole story history
underneath it. The retrieved memories and the rewritten summary did it again.

Everything fixed is emitted first now, and everything that moves goes after
the history, ordered least-volatile first — which is also where recency serves
it best, the reasoning that already put the emit reminder last. The three tail
sections that are last for their own reasons stay last. The moved sections are
still charged to the token budget; only their position changed.

Two smaller halves of the same problem. OpenRouter serves a model from
whichever upstream is free and each upstream holds its own cache, so a
deepseek model now names deepseek as its preferred upstream — a preference,
not a restriction, so a turn still runs if that upstream is down. And the
endpoint's usage block is read back off the response and kept per attempt, so
the hit rate shows up in Insights and the debug log instead of being assumed.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01DfMCsN1KBLsTqMkj5hSgrY

backend/app/routers/adventures.py

@@ -809,6 +809,10 @@ async def _generate_turn(
         snapshot["world_state"] = {"delta": delta, "report": ws_report, "state": new_world_state}
 
     snapshot["raw_output"] = raw_output
+    # What the endpoint says the call cost, including how much of the prompt it
+    # read from cache rather than billed in full. Recorded per attempt, beside
+    # the prompt it priced.
+    snapshot["usage"] = provider.last_usage
 
     reasoning = "".join(reasoning_chunks).strip() or None
     ai_action = models.Action(