feat: integrate TTS voice handling, improve loading UX, and stabilize runtime dependencies

[Afdaan]

Overview

This PR introduces improvements to Text-to-Speech (TTS) handling, enhances loading animations, and stabilizes runtime behavior across multiple components.

Merged from feat/voice into development.

---

Key Changes

Voice & TTS

• Added voice_helpers module for improved TTS handling
• Integrated !tts command handling
• Improved support for:
  • Replied voice messages
  • Group-based TTS triggers
• Simplified TTS dispatch flow
• TTS processing is now delegated to a separate microservice:
  • https://github.com/Afdaan/Alya-TTS
  • Designed to offload processing and reduce resource usage on the main bot service

---

Loading Animation Improvements

• Added timeout handling to prevent stuck animations
• Improved error handling during loading states
• Tweaked animation intervals and punctuation
• Ignored non-critical "not found" errors during animation

---

System & Stability

• Improved ChatActionSender error handling
• Removed redundant send_voice_reply call from VoiceHandler
• Simplified loading message flow

---

Notes

• Improves responsiveness and reliability of TTS features
• Reduces runtime errors during async operations
• Uses microservice-based architecture for better scalability

---

Impact

• More reliable and flexible TTS handling
• Reduced load on main bot service
• Better user experience with improved loading feedback
• Cleaner deployment pipeline and dependency management