v0.1.8: two-Spark pipeline + kept-warm serving

- Secondary-Spark models now work in air-gapped mode: the LiteLLM proxy is
  dual-homed onto the default bridge (docker network connect) to reach the
  secondary's published vLLM port; grader containers stay on the --internal
  network with zero egress. The head-only enforcement is replaced by a
  secondary-configured check.
- Extraction runs in parallel with grading when the extractor's model and
  every grader model in the wave sit on different Sparks (separate GPUs).
- Keep-warm: single-wave jobs no longer tear the wave down between decks
  (was a ~6-min 31B reload per deck); a kept-warm wave that fails preflight
  is restarted once. Adjudicator reuses the live wave when its model is
  already serving instead of cycling the shared proxy.
- clear_resident_containers (preJobStopContainers) now stops names on every
  configured Spark; health() reports containers on both Sparks.
- Verified with a mocked dry-run of the full job loop (3 decks: one
  bring-up, zero mid-job teardowns, parallel overlap, stale-wave restart).

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

startos/versions/index.ts

@@ -7,9 +7,10 @@ import { v_0_1_4 } from './v_0_1_4'
 import { v_0_1_5 } from './v_0_1_5'
 import { v_0_1_6 } from './v_0_1_6'
 import { v_0_1_7 } from './v_0_1_7'
+import { v_0_1_8 } from './v_0_1_8'
 
 /** The current version MUST be the first argument (`current`). */
 export const versions = VersionGraph.of({
-  current: v_0_1_7,
-  other: [v_0_1_6, v_0_1_5, v_0_1_4, v_0_1_3, v_0_1_2, v_0_1_1, v_0_1_0],
+  current: v_0_1_8,
+  other: [v_0_1_7, v_0_1_6, v_0_1_5, v_0_1_4, v_0_1_3, v_0_1_2, v_0_1_1, v_0_1_0],
 })