fix: block ALL LS follow-up requests across connections

Move the in-flight blocking check to the top of the LLM request flow,
BEFORE request modification. This catches follow-ups on ALL connections
(the LS opens multiple parallel TLS connections). Only the very first
modified request reaches Google — all others get fake STOP responses.

Previously, each new connection independently allowed one request
through before blocking, letting 4-5 requests leak per turn.

src/constants.rs

@@ -115,9 +115,7 @@ fn detect_versions() -> DetectedVersions {
     const FALLBACK_CLIENT: &str = "1.16.5";
 
     let Some(install_dir) = find_install_dir() else {
-        tracing::warn!(
-            "Could not find Antigravity install — using fallback versions"
-        );
+        tracing::warn!("Could not find Antigravity install — using fallback versions");
         return DetectedVersions {
             antigravity: FALLBACK_ANTIGRAVITY.to_string(),
             chrome: FALLBACK_CHROME.to_string(),