test: clean up the paid eval lane (B1-B4, B6, B7)

- B1: delete paid files that assert nothing or cannot pass meaningfully:
  skill-llm-eval-spec and skill-e2e-spec-execute (test.todo), gemini-e2e
  (+ gemini-session-runner; no gemini CLI in CI), ship-idempotency (red
  since v1.63), the two opus-4-7 *-sonnet overlay wrappers, conductor-prose
  (+ its source-evaluation replay), codex-e2e-plan-format; drop their keys,
  scripts and census rows.
- B2: skill-llm-eval grades browse/sections/command-list.md with one union
  judge that also carries the baseline score pin; regression-vs-baseline
  deleted (paid run: pass, c4/c4/a4).
- B3: memory-pipeline, ios-qa, ios-qa-swift-build and plan-tune-cathedral
  make no model calls; renamed out of the paid glob so they run on every
  PR. Swift builds need GSTACK_TEST_SWIFT=1; device stub deleted.
- B4: codex-e2e*, outside-voice, aside and ios-device cannot run in the CI
  image; excluded from the weekly lane with a tracked re-entry condition.
- B6: fold opus-47's negative routing controls into skill-routing-e2e
  journey-negatives (paid run: 3/3 unrouted) and delete the file.
- B7: delete the never-green brain-privacy-gate eval; a free
  gstack-skill-start test now proves consent precedes artifacts egress.

test/eng-finding-retry-budget.test.ts

@@ -127,16 +127,16 @@ test('live periodic census fits the declared CI wall including setup', () => {
     return paidShardWallUpperBoundMs(files, workers);
   });
   expect(Math.max(...walls) + 20 * 60_000).toBeLessThanOrEqual(periodicJob['timeout-minutes'] * 60_000);
-  expect(m.entries.filter(e => e.status === 'planned')).toHaveLength(100);
+  expect(m.entries.filter(e => e.status === 'planned')).toHaveLength(82);
   const overlays = m.entries.filter(e => e.status === 'planned' && e.slice === periodicSliceCount - 1);
-  expect(overlays).toHaveLength(6);
+  expect(overlays).toHaveLength(4);
   expect(overlays.every(e => isOverlayTestFile(e.file))).toBe(true);
   expect(m.entries.filter(e => e.status === 'planned' && e.slice === periodicSliceCount).map(e => e.file)).toEqual([AUTOPLAN_CHAIN_BUDGET.file]);
 });
 
 test('registered allocation is deterministic and preserves every discovered file', () => {
   const files = collectPaidTestFiles();
-  expect(files).toHaveLength(119);
+  expect(files).toHaveLength(105);
   const m = livePlan(files);
   expect(livePlan([...files].reverse())).toEqual(m);
   expect(m.entries.map(e => e.file).sort()).toEqual([...files].sort());
@@ -175,7 +175,7 @@ test('current detach supervision covers the live-census floor', () => {
   const floor = Math.ceil((Math.ceil(files.length / DEFAULT_JOBS) * DEFAULT_SHARD_TIMEOUT_MS + excess) / 1000 * 1.05);
   const pkg = JSON.parse(fs.readFileSync(path.join(import.meta.dir, '../package.json'), 'utf8'));
   const configured = Number(pkg.scripts['eval:bg:periodic'].match(/--timeout\s+(\d+)/)[1]);
-  expect(floor).toBe(65541);
+  expect(floor).toBe(57330);
   expect(configured).toBeGreaterThanOrEqual(floor);
   expect(pkg.scripts['eval:bg:gate']).toContain('--timeout 36000');
 });