Merge branch 'main' of github.com:MingLu0/SummarizerBackend

feat: Implement V3 Web Scraping + Summarization API

Merge pull request #1 from MingLu0/add-claude-github-actions-1762414221890

"Claude Code Review workflow"

"Claude PR Assistant workflow"

Update v2 API and HF streaming summarizer improvements

Improve V2 summarization: adaptive tokens, recursive summarization, better defaults

Comment out asyncio.sleep for better streaming performance

Document HF V2 summary length improvements in FAILED_TO_LEARN.MD

Enhance HF V2 summaries: increase token limits, improve length parameters

Improve V2 summarization prompts for more detailed output

Add model ID logging to HF streaming service

Fix HF streaming beam expansion issue

Enforce singleton batch for HF streaming (fix TextStreamer batch size 1 error)

Fix BatchEncoding normalization and attention_mask creation

Harden HF V2 streaming + full traceback logging

Fix HF streaming crash: enforce batch size = 1 for TextIteratorStreamer

Fix TextStreamer batch size error in V2 API

Add support for sshleifer/distilbart-cnn-6-6 model for V2 API

docs: update README with live deployment status and working URLs

fix(spaces): add FastAPI root_path for HF proxy, ensure 0.0.0.0:7860, verify v2 paths

Improve HuggingFace model error handling and startup robustness

Force HuggingFace Spaces restart - T5 compatibility fix

Fix T5 model compatibility in HuggingFace streaming service

Optimize V2 API for HuggingFace Spaces with t5-small

Add V2 API with HuggingFace streaming support

Fix all remaining test failures - 100% test success

Fix transformers service to be optional to prevent import errors

feat: Add auto-deploy setup for GitHub + Hugging Face

feat: Add Transformers pipeline endpoint for 80% faster summarization

feat: Update README for Hugging Face Spaces deployment

feat: Add Ollama model warmup for 35% performance improvement

docs: Update README.md with streaming endpoint documentation

feat: Add text streaming support with SSE

docs: Add Hugging Face Spaces live demo URL to README

docs: Update README.md with accurate model and configuration info

Document 504 Gateway Timeout fix learnings

Fix 504 Gateway Timeout: Increase timeout values

Optimize latency and performance for Hugging Face Spaces

Switch to llama3.2:1b base model for better availability

Force restart to apply llama3.2:1b-instruct configuration

Fix storage limit issue by switching to llama3.2:1b-instruct

Trigger restart for llama3.2:3b-instruct model

Switch to llama3.2:3b-instruct model for faster performance

Trigger restart for model update

Switch to mistral:7b-instruct model for better instruction following

Fix URL normalization and improve error handling in Ollama service

Add debug config endpoint to check environment variables

Add debug logging to health check

Add debug logging to track Ollama URL construction