Make the Scraper Reliable

[dragosrotaru]

• [ ] Implement partial scraping - recent members only, detect field change frequency, prioritize based on group size, history of growth
• [ ] don't do batch jobs, let the scraper run as a chron job or background task
• [ ] use counts vs list.length to detect issues
• [ ] save stderr and scraper parameters
• [x] connect to existing browser or save session (no relogin)
• [x] make auto-scrolling work by waiting for network idle
• [x] create phony public and private groups for integration tests
• [ ] deploy scraper on multiple personal computers (Scraping@Home V0.0.1)
• [ ] use exponential backoffs