Content inspection and archive walks on seeding nodes (relay)

Inspection is split into prepare (database), probe (lib/content/probe:
downloads, decoding and parsing, no database) and apply (database).
worker:collection runs all three; seed:node with SEED_INSPECTORS leases
jobs from the same queues through the seed API (BullMQ manual leases),
fetches articles through the seed API and posts the probes back.

A node now needs only SEED_URL and SEED_TOKEN: the settings the app
requires but a node never uses get placeholders for seed:node.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>

tests/unit/rar.spec.ts

@@ -325,7 +325,7 @@ test.group('rar / archive walks', () => {
   })
 
   test('finds segments from their yEnc offsets', async ({ assert }) => {
-    const { SegmentReader } = await import('#services/ArchiveWalkService')
+    const { SegmentReader } = await import('#lib/rar/index')
     const set = rar5Set()
     const size = 700
     const articles = new Map<string, Buffer>()
@@ -336,7 +336,7 @@ test.group('rar / archive walks', () => {
         articles.set(id, yencArticle(`x.part${v + 1}.rar`, volume, i, size))
         segments.push({ number: i + 1, id, size: 0 })
       }
-      return { id: v + 1, segments }
+      return segments
     })
     const reader = new SegmentReader(files, 100, async (id) => articles.get(id)!)
     const walk = await walkRar(reader)