fix: answer telegram callback queries immediately (#1349) (thanks @siddhantjain)

Telegram retries callback queries if they aren't acknowledged quickly.
Previously, answerCallbackQuery was called in a finally block AFTER
processing, which could take several seconds for agent responses.

This change moves answerCallbackQuery to immediately after basic
validation, before any processing begins. This prevents Telegram
from sending duplicate callbacks while the agent is thinking.

Fixes duplicate callback handling when agent processing is slow.

CHANGELOG.md

@@ -24,6 +24,7 @@ Docs: https://docs.clawd.bot
 - UI: add copy-as-markdown with error feedback and drop legacy list view. (#1345) — thanks @bradleypriest.
 ### Fixes
 - Discovery: shorten Bonjour DNS-SD service type to `_clawdbot-gw._tcp` and update discovery clients/docs.
+- Telegram: answer callback queries immediately to prevent retries. (#1349) — thanks @siddhantjain.
 - Agents: preserve subagent announce thread/topic routing + queued replies across channels. (#1241) — thanks @gnarco.
 - Agents: avoid treating timeout errors with "aborted" messages as user aborts, so model fallback still runs.
 - Diagnostics: export OTLP logs, correct queue depth tracking, and document message-flow telemetry.

src/telegram/bot-handlers.ts

@@ -179,6 +179,8 @@ export const registerTelegramHandlers = ({
     const callback = ctx.callbackQuery;
     if (!callback) return;
     if (shouldSkipUpdate(ctx)) return;
+    // Answer immediately to prevent Telegram from retrying while we process
+    await bot.api.answerCallbackQuery(callback.id).catch(() => {});
     try {
       const data = (callback.data ?? "").trim();
       const callbackMessage = callback.message;
@@ -323,8 +325,6 @@ export const registerTelegramHandlers = ({
       });
     } catch (err) {
       runtime.error?.(danger(`callback handler failed: ${String(err)}`));
-    } finally {
-      await bot.api.answerCallbackQuery(callback.id).catch(() => {});
     }
   });