feat(podcast): add pre-estimate endpoint, enhance cost estimator with multi-model support, cleanup alpha pricing seeding

- Add POST /podcast/pre-estimate endpoint for cost estimation before analysis
- Enhance cost_estimator.py with multi-model support (gemini, audio, voice clone, image, video)
- Add detailed cost breakdown (llm, audio, media costs + per-phase breakdown)
- Remove redundant pricing seeding from init_alpha_subscription_tiers.py
- Add SSOT pricing via PricingService.initialize_default_pricing()
- Update TopicUrlInput tooltip to show estimate details
- Add debug logging for pricing seeding and pre-estimate
- Clean up verbose podcast mode debug logs in app.py

backend/services/llm_providers/audio_to_text_generation/gemini_audio_text.py

@@ -67,10 +67,11 @@ import sys
 from pathlib import Path
 import google.genai as genai
 from google.genai import types
-
+from dotenv import load_dotenv
 
 from loguru import logger
 from utils.logger_utils import get_service_logger
+from services.api_key_manager import APIKeyManager
 
 # Use service-specific logger to avoid conflicts
 logger = get_service_logger("gemini_audio_text")