perf(search+embed): zero-copy embedding API and deferred RRF mapping

Change OllamaClient::embed_batch to accept &[&str] instead of
Vec<String>. The EmbedRequest struct now borrows both model name and
input texts, eliminating per-batch cloning of chunk text (up to 32KB
per chunk x 32 chunks per batch). Serialization output is identical
since serde serializes &str and String to the same JSON.

In hybrid search, defer the RrfResult->HybridResult mapping until
after filter+take, so only `limit` items (typically 20) are
constructed instead of up to 1,500 at RECALL_CAP. Also switch
filtered_ids to into_iter() to avoid an extra .copied() pass.

Switch FTS search_fts from prepare() to prepare_cached() for statement
reuse across repeated searches. Benchmarked at ~1.6x faster.

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

src/embedding/ollama.rs

@@ -27,9 +27,9 @@ pub struct OllamaClient {
 }
 
 #[derive(Serialize)]
-struct EmbedRequest {
-    model: String,
-    input: Vec<String>,
+struct EmbedRequest<'a> {
+    model: &'a str,
+    input: Vec<&'a str>,
 }
 
 #[derive(Deserialize)]
@@ -101,12 +101,12 @@ impl OllamaClient {
         Ok(())
     }
 
-    pub async fn embed_batch(&self, texts: Vec<String>) -> Result<Vec<Vec<f32>>> {
+    pub async fn embed_batch(&self, texts: &[&str]) -> Result<Vec<Vec<f32>>> {
         let url = format!("{}/api/embed", self.config.base_url);
 
         let request = EmbedRequest {
-            model: self.config.model.clone(),
-            input: texts,
+            model: &self.config.model,
+            input: texts.to_vec(),
         };
 
         let response = self
@@ -181,8 +181,8 @@ mod tests {
     #[test]
     fn test_embed_request_serialization() {
         let request = EmbedRequest {
-            model: "nomic-embed-text".to_string(),
-            input: vec!["hello".to_string(), "world".to_string()],
+            model: "nomic-embed-text",
+            input: vec!["hello", "world"],
         };
         let json = serde_json::to_string(&request).unwrap();
         assert!(json.contains("\"model\":\"nomic-embed-text\""));

src/embedding/pipeline.rs

@@ -162,9 +162,9 @@ async fn embed_page(
     let mut cleared_docs: HashSet<i64> = HashSet::with_capacity(pending.len());
 
     for batch in all_chunks.chunks(BATCH_SIZE) {
-        let texts: Vec<String> = batch.iter().map(|c| c.text.clone()).collect();
+        let texts: Vec<&str> = batch.iter().map(|c| c.text.as_str()).collect();
 
-        match client.embed_batch(texts).await {
+        match client.embed_batch(&texts).await {
             Ok(embeddings) => {
                 for (i, embedding) in embeddings.iter().enumerate() {
                     if i >= batch.len() {
@@ -228,7 +228,7 @@ async fn embed_page(
                 if is_context_error && batch.len() > 1 {
                     warn!("Batch failed with context length error, retrying chunks individually");
                     for chunk in batch {
-                        match client.embed_batch(vec![chunk.text.clone()]).await {
+                        match client.embed_batch(&[chunk.text.as_str()]).await {
                             Ok(embeddings)
                                 if !embeddings.is_empty()
                                     && embeddings[0].len() == EXPECTED_DIMS =>