perf(llama-server): enable flash attention

Cuts prefill time with no accuracy cost, per markaicode.com's
llama.cpp timeout writeup — directly helps the prefill-vs-stream-idle
contention issue fixed in the prior two commits.

Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>

docker-compose.yml

@@ -20,6 +20,7 @@ services:
       --n-gpu-layers ${LLAMA_GPU_LAYERS:-999}
       --ctx-size ${LLAMA_CTX_SIZE:-131072}
       --parallel ${LLAMA_PARALLEL:-2}
+      --flash-attn on
       --jinja
     # No published host port: llama-server is reached only via the omniroute
     # gateway on the ai-stack docker network now — see issue #15. Its