feat(webui): persist crawler configuration and expose crawl limits

[yaoziqin2020]

Summary

• expose post and per-post comment limits in the WebUI and pass them to crawler start requests
• add explicit persistence for the current safe WebUI configuration while excluding cookies
• split login and crawl settings into aligned cards and move the save action into the header
• prevent header control labels from being truncated and keep the original terminal banner frame closed

Why

The WebUI previously restored default values after reopening and did not expose the crawler's post and comment limits. This made repeated local research runs require the same setup each time and forced limit changes to be made outside the UI.

Behavior

• edits affect the current crawler configuration immediately
• clicking Save all configuration controls what is restored on the next page load
• unsaved edits do not overwrite the last saved configuration
• cookies are intentionally excluded from local persistence
• the default limits remain 20 posts and 10 comments per post

Validation

• npm run build
• git diff --check
• browser QA for card alignment, input editing, save feedback, saved-value restoration, unsaved-value restoration, header labels, and terminal banner rendering

The WebUI package does not currently define an automated test script.