feat: retry channel startup on transient failures (max 5 attempts, exponential backoff)

Previously, a single transient network error during channel startup
(e.g., proxy blip causing Telegram get_me() to fail) would leave the
channel permanently dead until the next gateway restart.

Now _start_channel retries up to 5 times with exponential backoff
starting at 2s (2s, 4s, 8s, 16s, 32s = ~62s total max wait).

nanobot/channels/manager.py

@@ -330,19 +330,40 @@ class ChannelManager:
         value = getattr(section, key, None)
         return value if isinstance(value, bool) else default
 
+    # Retry settings for channel startup (transient failures like network blips)
+    _CHANNEL_START_MAX_RETRIES: int = 5
+    _CHANNEL_START_BASE_DELAY: float = 2.0  # seconds, doubled each retry
+
     async def _start_channel(self, name: str, channel: BaseChannel) -> None:
-        """Start a channel and log any exceptions."""
+        """Start a channel with retries for transient failures."""
         errors = getattr(self, "_channel_errors", None)
         if errors is None:
             errors = self._channel_errors = {}
         errors.pop(name, None)
-        try:
+
-            await channel.start()
+        last_exc: Exception | None = None
-        except asyncio.CancelledError:
+        for attempt in range(self._CHANNEL_START_MAX_RETRIES + 1):
-            raise
+            try:
-        except Exception:
+                await channel.start()
-            errors[name] = "Channel failed to start. Check gateway logs."
+                if attempt > 0:
-            logger.exception("Failed to start channel {}", name)
+                    logger.info("Channel {} started on attempt {}", name, attempt + 1)
+                return
+            except asyncio.CancelledError:
+                raise
+            except Exception as exc:
+                last_exc = exc
+                if attempt < self._CHANNEL_START_MAX_RETRIES:
+                    delay = self._CHANNEL_START_BASE_DELAY * (2 ** attempt)
+                    logger.warning(
+                        "Channel {} start attempt {}/{} failed: {}. Retrying in {:.0f}s...",
+                        name, attempt + 1, self._CHANNEL_START_MAX_RETRIES + 1,
+                        exc, delay,
+                    )
+                    await asyncio.sleep(delay)
+
+        errors[name] = "Channel failed to start. Check gateway logs."
+        logger.exception("Failed to start channel {} after {} attempts",
+                         name, self._CHANNEL_START_MAX_RETRIES + 1)
 
     def _start_channel_task(self, name: str, channel: BaseChannel) -> asyncio.Task:
         logger.info("Starting {} channel...", name)