Rebuild: first end-to-end runs; prefetch per chunk, executable rule, wire fixes

End to end on the workstation (coordinator:run, the API, edge:node, real
NNTP): a.b.e-book.french and a.b.cd.image.playstation2.dvdiso ranges.

- Prefetcher: every connection takes the next missing chunk (leased
  ranges first, in order); one connection per range (~26k lines/s) kept
  the worker waiting
- executable rule: releases of bare executables (malware floods named
  after films and software) are dropped; they were 92% of the catalog rows
  of the test ranges, 21% of the segments
- fragments per file (details once) and gzipped batch bodies; the JSON
  limit (1 MB by default) raised to 256 MB, empty strings kept (a rule
  count's key is '')
- release_chunks.batch_from is a bigint (boneless article numbers)
- tests get their own database; edge:bench --store times BatchStore:
  1.3 s for a boneless range (1.24 ms per release chunk)

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

commands/edge_node.ts

@@ -1,7 +1,7 @@
 import { BaseCommand, flags } from '@adonisjs/core/ace'
 import type { CommandOptions } from '@adonisjs/core/types/ace'
 import { setTimeout as sleep } from 'node:timers/promises'
-import { unzipSync } from 'node:zlib'
+import { gzipSync, unzipSync } from 'node:zlib'
 import { parseOverviews } from '#lib/nntp/index'
 import { RangeIndexer, decodeFrames, toWire, type EdgeBatch, type Lease } from '#lib/edge/index'
 
@@ -68,10 +68,13 @@ export default class EdgeNode extends BaseCommand {
     const { range, rules } = lease
     const started = Date.now()
     const indexer = new RangeIndexer({ group: range.group, start: range.start, end: range.end, rules })
-    const totals = { headers: 0, releases: 0, candidates: 0 }
+    const totals = { headers: 0, releases: 0, candidates: 0, bytes: 0, postMs: 0 }
     let chunks = 0
     const post = async (batch: EdgeBatch) => {
+      const at = Date.now()
       const stored = await api.batch(range.id, batch)
+      totals.postMs += Date.now() - at
+      totals.bytes += stored.bytes
       totals.headers += batch.headers
       totals.releases += stored.releases
       totals.candidates += stored.candidates
@@ -94,7 +97,8 @@ export default class EdgeNode extends BaseCommand {
       const seconds = (Date.now() - started) / 1000
       this.logger.info(
         `${range.group} ${range.start}–${range.end}: ${totals.headers.toLocaleString('en')} headers in ${seconds.toFixed(0)} s ` +
-          `(${Math.round(totals.headers / seconds).toLocaleString('en')}/s), ${totals.releases} release chunks, ${totals.candidates} probe candidates.`
+          `(${Math.round(totals.headers / seconds).toLocaleString('en')}/s), ${totals.releases} release chunks, ${totals.candidates} probe candidates; ` +
+          `posted ${(totals.bytes / 1e6).toFixed(1)} MB in ${(totals.postMs / 1000).toFixed(1)} s.`
       )
     } catch (error: any) {
       if (error instanceof LeaseLost) {
@@ -144,10 +148,12 @@ class Api {
     return response.body as unknown as AsyncIterable<Uint8Array>
   }
 
-  async batch(rangeId: number, batch: EdgeBatch): Promise<{ releases: number; candidates: number }> {
-    const response = await this.request(`ranges/${rangeId}/batches`, { method: 'POST', body: JSON.stringify(toWire(batch)) })
+  async batch(rangeId: number, batch: EdgeBatch): Promise<{ releases: number; candidates: number; bytes: number }> {
+    // Gzipped: chunk payloads are compressed already, the rest (names, fragments, candidates) compresses well.
+    const body = gzipSync(JSON.stringify(toWire(batch)), { level: 1 })
+    const response = await this.request(`ranges/${rangeId}/batches`, { method: 'POST', body, headers: { 'Content-Encoding': 'gzip' } })
     if (!response.ok) throw new Error(`batch: HTTP ${response.status}`)
-    return response.json() as Promise<{ releases: number; candidates: number }>
+    return { ...((await response.json()) as { releases: number; candidates: number }), bytes: body.length }
   }
 
   async call(rangeId: number, action: 'finish' | 'release' | 'heartbeat') {