perf: eliminate unnecessary clones and pre-allocate collections

Three micro-optimizations with zero behavioral change:

1. timeline_collect.rs: Reorder format!() before enum construction so
   the owned String moves into the variant directly, eliminating
   .clone() on state, label, and milestone strings in StateChanged,
   LabelAdded/Removed, and MilestoneSet/Removed event paths.

2. pipeline.rs: Use Arc<str> for doc_hash shared across a document's
   chunks instead of cloning the full String per chunk. Also remove
   redundant embed_buf.reserve() since extend_from_slice already
   handles growth and the buffer is reused across iterations.

3. rrf.rs: Pre-allocate HashMap with combined vector+fts result count
   via with_capacity() to avoid rehashing during RRF score accumulation.

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

src/embedding/pipeline.rs

@@ -1,4 +1,5 @@
 use std::collections::{HashMap, HashSet};
+use std::sync::Arc;
 
 use futures::future::join_all;
 use rusqlite::Connection;
@@ -28,7 +29,7 @@ struct ChunkWork {
     doc_id: i64,
     chunk_index: usize,
     total_chunks: usize,
-    doc_hash: String,
+    doc_hash: Arc<str>,
     chunk_hash: String,
     text: String,
 }
@@ -212,12 +213,13 @@ async fn embed_page(
 
         chunks_needed.insert(doc.document_id, total_chunks);
 
+        let doc_hash: Arc<str> = Arc::from(doc.content_hash.as_str());
         for (chunk_index, text) in chunks {
             all_chunks.push(ChunkWork {
                 doc_id: doc.document_id,
                 chunk_index,
                 total_chunks,
-                doc_hash: doc.content_hash.clone(),
+                doc_hash: Arc::clone(&doc_hash),
                 chunk_hash: sha256_hash(&text),
                 text,
             });
@@ -501,7 +503,6 @@ fn store_embedding(
     let rowid = encode_rowid(doc_id, chunk_index as i64);
 
     embed_buf.clear();
-    embed_buf.reserve(embedding.len() * 4);
     for f in embedding {
         embed_buf.extend_from_slice(&f.to_le_bytes());
     }