Cache feed/llms.txt bodies, serve with content-hash ETags

Rendering every article per request is expensive. Bodies are now
RAM-cached keyed by the public base URL and cleared from
_invalidate_pages (the same hook that drops the page render cache on
any content/settings write). Responses carry a blake3 content-hash
ETag and answer 304, so polling feed readers revalidate cheaply;
matching page behavior, 304 revalidations are not recorded in
analytics.

pagerite/state.py

@@ -171,11 +171,16 @@ _render_gen = 0
 
 
 def _invalidate_pages() -> None:
-    """Drop cached page bodies and bump the render generation (ETags);
-    any content change also re-runs translation dispatch."""
+    """Drop cached page bodies (and the feed/llms.txt export bodies) and
+    bump the render generation (ETags); any content change also re-runs
+    translation dispatch."""
     global _render_gen
     _render_gen += 1
     _cached_body.cache_clear()
+    # Local import: pagerite.feeds imports this module.
+    from pagerite import feeds
+
+    feeds._cached_feed.cache_clear()
     dispatcher.schedule()