Rebuild: coordinator (planner, NNTP pool, spool, prefetcher, leases, rule sets), node API and the edge:node worker

- groups:add / ranges:plan / nodes:create / coordinator:run commands
- the node API: lease, the range's chunks as one framed stream from the
  spool, batches, heartbeat, finish, release (node guard, token per node)
- edge:node: needs COORDINATOR_URL and NODE_TOKEN only (placeholders for
  the rest), posts a batch every 10 chunks, gives its range back on SIGTERM
- releases are gathered per batch: 1.04 chunk rows per release at 10
  chunks per batch (1.35 when built per XOVER chunk)
- partitions are checked per batch instead of remembered

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

lib/edge/range_indexer.ts

@@ -143,6 +143,7 @@ export class RangeIndexer {
 
   /** What was found since the last take, and a fresh batch. */
   take(): EdgeBatch {
+    this.judgeReleases()
     const out = this.batch
     out.counts = [...this.counts.values()]
     this.counts.clear()
@@ -154,14 +155,22 @@ export class RangeIndexer {
     return this.aggregator.openFiles
   }
 
+  /** Completed files, gathered per release until the batch is taken: one chunk per release per batch. */
+  private completed = new Map<string, AggregatedFile<IngestSegment>[]>()
+
   private files(files: AggregatedFile<IngestSegment>[]) {
-    const byRelease = new Map<string, AggregatedFile<IngestSegment>[]>()
     for (const file of files) {
       const key = releaseKeyOf(file.segments[0])
-      const list = byRelease.get(key)
+      const list = this.completed.get(key)
       if (list) list.push(file)
-      else byRelease.set(key, [file])
+      else this.completed.set(key, [file])
     }
+  }
+
+  /** Judges and encodes the releases completed in this batch. */
+  private judgeReleases() {
+    const byRelease = this.completed
+    this.completed = new Map()
     for (const [key, list] of byRelease) {
       const verdict = this.rules.judge(
         key,