Introduce RabbitMQ consumer pool for parallel processing

Enhanced RabbitMQ email processing by introducing a `SendingEmailConsumerPool` to enable the competing consumers pattern. Each consumer operates on its own channel, improving scalability and thread safety.

- Added `SendingEmailConsumerPool` to manage multiple consumers.
- Updated `DependencyInjection` to register the consumer pool.
- Refactored `SendingEmailConsumer` for better logging and error handling.
- Updated `AsyncInitWorker` to initialize the consumer pool.
- Added `ConsumerConcurrency` to RabbitMQ configuration.
- Improved error handling in `LimilabsEmailService` with detailed SMTP error messages.

src/infrastructure/DigitalData.MessagingService.RabbitMQ/RabbitMqConfiguration.cs

@@ -75,5 +75,11 @@ namespace DigitalData.MessagingService.RabbitMQ
         /// Routing key used to bind <see cref="DlqQueueName"/> to <see cref="DlqExchangeName"/>.
         /// </summary>
         public string DlqRoutingKey { get; set; } = "email.outbox.dlq";
+
+        /// <summary>
+        /// Maximum number of email messages processed concurrently by the consumer.
+        /// Maps directly to RabbitMQ prefetchCount. Recommended: 3–5.
+        /// </summary>
+        public ushort ConsumerConcurrency { get; set; } = 5;
     }
 }