llama-swap: drop ttl: 600 from qwen3.6 abliterated, heretic, 27b — keep loaded across idle

The base qwen3.6-35-a3b is already ttl: 0 via the `pinned` group.
The three other Qwen 3.6 variants (abliterated, heretic, 27b) had
ttl: 600 → llama-swap auto-unloaded them after 10 min idle, costing
the next request a full reload (~5-15s). Removed so they stay loaded
once warm. Still get evicted by the normal swap when another
non-pinned model is requested — these aren't joining the pinned group,
just losing their idle-unload timer.

stacks/llama-swap/conf/config.yaml

@@ -151,7 +151,8 @@ models:
   "qwen3.6-35-a3b-abliterated":
     name: "Qwen 3.6 35B-A3B Abliterated i1-Q6_K"
     description: "Qwen3.6 MoE, 3B active of 35B. mradermacher abliterated imatrix Q6_K (~27GB)."
-    ttl: 600
+    # No ttl — stays loaded across idle. Still gets evicted by the
+    # normal swap when another non-pinned model is requested.
     cmd: |
       /app/llama-server
       --context-shift
@@ -173,7 +174,8 @@ models:
   "qwen3.6-35-a3b-heretic":
     name: "Qwen 3.6 35B-A3B Uncensored Heretic Q6_K (vision)"
     description: "Qwen3.6 MoE, 3B active of 35B. llmfan46 uncensored 'heretic' Q6_K (~28GB) + auto-loaded BF16 mmproj for vision."
-    ttl: 600
+    # No ttl — stays loaded across idle. Still gets evicted by the
+    # normal swap when another non-pinned model is requested.
     cmd: |
       /app/llama-server
       --context-shift
@@ -195,7 +197,8 @@ models:
   "qwen3.6-27b":
     name: "Qwen 3.6 27B UD-Q6_K_XL (vision)"
     description: "Qwen3.6 dense 27B. Unsloth dynamic Q6_K_XL (~26GB) + auto-loaded BF16 mmproj for vision."
-    ttl: 600
+    # No ttl — stays loaded across idle. Still gets evicted by the
+    # normal swap when another non-pinned model is requested.
     cmd: |
       /app/llama-server
       --context-shift