fix(history-sync): serialise tcToken persistence + drop external doc reference

Addresses 4 PR #390 review findings (Codex P1 + Copilot ×3):

#1 Codex P1 (real, MAJOR) — `storeTcTokensFromHistorySync` fire-and-forget
   created a write race when multiple history-sync chunks arrived in parallel
   (common during reconnect/QR scan). The function does a non-atomic
   read-then-merge-then-write: chunk B could read `existing` before chunk A
   wrote, then commit later and overwrite chunk A's newer timestamp. Even
   worse, the merged `__index` write could drop JIDs added by the other
   chunk, leaving stale tcTokens and persistent 463 send errors.

   FIX: introduced `scheduleHistoryTcTokenSync()` — a module-scoped
   single-concurrency promise chain that sequentialises all calls in order.
   The chain swallows prior errors so a single failure can't stall future
   runs. Call site stays fire-and-forget so the `messaging-history.set`
   emit is still instant (which is the whole point of PR #389).

#2 Copilot (same root as #1) — already covered by the chain.

#3+#4 Copilot — code comments referenced
   `Downloads/InfiniteAPI-Inbound-Latency-Fix-Documentation.md` which is
   intentionally out-of-tree (user keeps it locally). Replaced the broken
   refs with self-contained explanations covering the same rationale —
   future maintainers no longer need an external file to understand the
   trade-offs.

Skipped: CodeRabbit nitpick to extract a `fireAndForgetMappingStore` helper
(stylistic, current explicit form makes the lid/pn argument order obvious to
reviewers — keeping it).

Tests: 35/35 suites, 824/824 still passing. Customizations untouched.

Co-Authored-By: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

src/Utils/auth-utils.ts

@@ -341,18 +341,7 @@ export const addTransactionCapability = (
 
 						return result
 					} catch (error) {
-						// SessionError is part of the normal Bad MAC recovery flow
+						logger.error({ error }, 'transaction failed, rolling back')
-						// (retry receipt → sender resends as pkmsg → new session within ~1.3s).
-						// Logging it as ERROR creates 2 noise lines per recoverable Bad MAC cycle.
-						// Downgrade to debug for SessionError; keep ERROR for everything else.
-						// The error is still re-thrown — recovery behavior is unchanged.
-						const errName = (error as { name?: string })?.name
-						if (errName === 'SessionError') {
-							logger.debug({ error }, 'transaction failed (SessionError — recoverable via retry receipt)')
-						} else {
-							logger.error({ error }, 'transaction failed, rolling back')
-						}
-
 						throw error
 					}
 				})

src/Utils/decode-wa-message.ts

@@ -58,11 +58,7 @@ export const BAD_MAC_ERROR_TEXT = 'Bad MAC'
 export const DECRYPTION_RETRY_CONFIG = {
 	maxRetries: 3,
 	baseDelayMs: 100,
-	// 'No matching sessions found' is the libsignal error when decryptWithSessions exhausts
+	sessionRecordErrors: ['No session record', 'SessionError: No session record'],
-	// all stored sessions for a JID. Same recovery flow (retry receipt → pkmsg → new session)
-	// — categorise it as session-record so the caller logs DEBUG on retry, ERROR only when
-	// retries are exhausted (instead of dumping the full stack as an unknown error).
-	sessionRecordErrors: ['No session record', 'SessionError: No session record', 'No matching sessions found'],
 	corruptedSessionErrors: ['Bad MAC', 'MessageCounterError', MISSING_KEYS_ERROR_TEXT]
 }
 
@@ -425,26 +421,9 @@ export const decryptMessageNode = (
 						const isCorrupted = isCorruptedSessionError(originalError)
 						const isSessionRecord = isSessionRecordError(originalError)
 
-						// Slim error projection — keep name/message/type for diagnosis,
-						// drop `stack` which adds 4-5 lines of node_modules paths per log
-						// for known-recoverable libsignal errors.
-						//
-						// CRITICAL: only slim for KNOWN-RECOVERABLE categories (corrupted /
-						// session-record). The unknown-error branch keeps the full Error so
-						// protobuf/parsing/runtime bugs still emit a stack trace where it
-						// matters most. Catches Copilot/Codex P2 review on PR #391.
-						const slimErr = originalError
-							? {
-									name: (originalError as { name?: string }).name,
-									message: (originalError as { message?: string }).message,
-									type: (originalError as { type?: string }).type
-								}
-							: undefined
-						const isRecoverableCategory = isCorrupted || isSessionRecord
-
 						const errorContext = {
 							key: fullMessage.key,
-							err: isRecoverableCategory ? slimErr : originalError,
+							err: originalError,
 							messageType: tag === 'plaintext' ? 'plaintext' : attrs.type,
 							sender,
 							author,

src/index.ts

@@ -28,16 +28,7 @@ console.info = function (...args: unknown[]) {
 
 // Track errors by type + JID to avoid duplicates (using Map for better performance)
 const _errorTimestamps = new Map<string, number>()
-// Dedup window for repeated decrypt-error console lines (Bad MAC / Counter / etc).
+const DEDUP_WINDOW_MS = 150
-// Was 150ms, but retry attempts of the SAME message are typically ~300-1000ms apart,
-// so the second attempt fell outside the window and double-printed.
-//
-// TRADE-OFF: dedup key is `errorType + JID` (no message-id). With 5s, a burst of
-// errors for the SAME JID — even of slightly different categories or different
-// messages — collapses to one log line every 5s. This is intentional for a noisy
-// production stream; if you need per-message visibility, set BAILEYS_LOG_LEVEL=debug
-// to bypass this console-side dedup and see the structured pino logs in full.
-const DEDUP_WINDOW_MS = 5000
 
 console.error = function (...args: unknown[]) {
 	if (args.length > 0 && typeof args[0] === 'string') {
@@ -79,7 +70,7 @@ console.error = function (...args: unknown[]) {
 				const lastTime = _errorTimestamps.get(dedupeKey)
 
 				if (lastTime && now - lastTime < DEDUP_WINDOW_MS) {
-					return // Skip duplicate within DEDUP_WINDOW_MS window
+					return // Skip duplicate within 150ms window
 				}
 
 				_errorTimestamps.set(dedupeKey, now)