feat(memory): Add MMR re-ranking for search result diversity

Adds Maximal Marginal Relevance (MMR) re-ranking to hybrid search results.

- New mmr.ts with tokenization, Jaccard similarity, and MMR algorithm
- Integrated into mergeHybridResults() with optional mmr config
- 40 comprehensive tests covering edge cases and diversity behavior
- Configurable lambda parameter (default 0.7) to balance relevance vs diversity
- Updated CHANGELOG.md and memory docs

This helps avoid redundant results when multiple chunks contain similar content.

docs/concepts/memory.md

@@ -353,7 +353,6 @@ agents: {
 ```
 
 Tools:
-
 - `memory_search` — returns snippets with file + line ranges.
 - `memory_get` — read memory file content by path.
 
@@ -396,11 +395,11 @@ But it can be weak at exact, high-signal tokens:
 
 - IDs (`a828e60`, `b3b9895a…`)
 - code symbols (`memorySearch.query.hybrid`)
-- error strings (“sqlite-vec unavailable”)
+- error strings ("sqlite-vec unavailable")
 
 BM25 (full-text) is the opposite: strong at exact tokens, weaker at paraphrases.
 Hybrid search is the pragmatic middle ground: **use both retrieval signals** so you get
-good results for both “natural language” queries and “needle in a haystack” queries.
+good results for both "natural language" queries and "needle in a haystack" queries.
 
 #### How we merge results (the current design)
 
@@ -423,12 +422,28 @@ Notes:
 
 - `vectorWeight` + `textWeight` is normalized to 1.0 in config resolution, so weights behave as percentages.
 - If embeddings are unavailable (or the provider returns a zero-vector), we still run BM25 and return keyword matches.
-- If FTS5 can’t be created, we keep vector-only search (no hard failure).
+- If FTS5 can't be created, we keep vector-only search (no hard failure).
 
-This isn’t “IR-theory perfect”, but it’s simple, fast, and tends to improve recall/precision on real notes.
+This isn't "IR-theory perfect", but it's simple, fast, and tends to improve recall/precision on real notes.
 If we want to get fancier later, common next steps are Reciprocal Rank Fusion (RRF) or score normalization
 (min/max or z-score) before mixing.
 
+#### MMR re-ranking (diversity)
+
+When hybrid search returns results, multiple chunks may contain similar or overlapping content.
+**MMR (Maximal Marginal Relevance)** re-ranks the results to balance relevance with diversity,
+ensuring the top results aren't all saying the same thing.
+
+How it works:
+1. Results are scored by their original relevance (vector + BM25 weighted score).
+2. MMR iteratively selects results that maximize: `λ × relevance − (1−λ) × similarity_to_selected`.
+3. Already-selected results are penalized via Jaccard text similarity.
+
+The `lambda` parameter controls the trade-off:
+- `lambda = 1.0` → pure relevance (no diversity penalty)
+- `lambda = 0.0` → maximum diversity (ignores relevance)
+- Default: `0.7` (balanced, slight relevance bias)
+
 Config:
 
 ```json5
@@ -440,7 +455,11 @@ agents: {
           enabled: true,
           vectorWeight: 0.7,
           textWeight: 0.3,
-          candidateMultiplier: 4
+          candidateMultiplier: 4,
+          mmr: {
+            enabled: true,
+            lambda: 0.7
+          }
         }
       }
     }