Add user slug to transcription jobs

The `TranscribeJob` struct now includes a `user_slug` field. This is
necessary for the transcription worker to look up per-user settings such
as hotwords and initial prompts. The `collect_pending` function has been
updated to return user slugs along with audio paths, and the recovery
scan now enqueues jobs with the correct user slug.

server/src/routes/upload.rs

@@ -86,7 +86,10 @@ pub async fn handle_upload(
 
     // Enqueue transcription. Failure here is non-fatal: on restart the recovery
     // scan will re-enqueue any .m4a without a .transcript.txt sibling.
-    match transcribe_tx.try_send(TranscribeJob { audio_path: file_path.clone() }) {
+    match transcribe_tx.try_send(TranscribeJob {
+        audio_path: file_path.clone(),
+        user_slug: user.slug.clone(),
+    }) {
         Ok(()) => {}
         Err(TrySendError::Full(_)) => {
             warn!(file = %file_path.display(), "Transcription queue full; will be picked up on restart");