Add author detail, idea detail, and gap-draft reverse link pages

- Author detail page (/authors/<person_id>): shows author info, all drafts
  with ratings, and co-authors with shared draft counts. Public route.
- Idea detail page (/ideas/<idea_id>): shows idea metadata, source draft,
  and top-5 most similar ideas via embedding cosine similarity. Admin route.
- Gap detail page: added "Related Drafts" section that finds drafts by
  extracting draft names from evidence text and searching by topic keywords.
- Updated author links across templates to use /authors/<person_id> URLs.
- Added DB methods: get_author_by_id, get_author_drafts, get_coauthors.
- Extended top_authors to include person_id (5th tuple element).

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

src/webui/data/analysis.py

@@ -103,6 +103,78 @@ def get_ideas_by_type(db: Database) -> dict:
         "ideas": all_ideas,
     }
 
+def get_idea_detail(db: Database, idea_id: int) -> dict | None:
+    """Return a single idea with source draft info and similar ideas."""
+    row = db.conn.execute("SELECT * FROM ideas WHERE id = ?", (idea_id,)).fetchone()
+    if not row:
+        return None
+
+    idea = {
+        "id": row["id"],
+        "title": row["title"],
+        "description": row["description"],
+        "type": row["idea_type"],
+        "draft_name": row["draft_name"],
+        "novelty_score": row["novelty_score"],
+    }
+
+    # Get source draft info
+    draft = db.get_draft(row["draft_name"])
+    if draft:
+        idea["draft_title"] = draft.title
+        idea["draft_date"] = draft.date
+
+    # Get category from ratings
+    rated = db.drafts_with_ratings(limit=2000)
+    for d, r in rated:
+        if d.name == row["draft_name"]:
+            idea["categories"] = r.categories
+            break
+
+    # Find similar ideas using embeddings
+    similar = []
+    emb_row = db.conn.execute(
+        "SELECT vector FROM idea_embeddings WHERE idea_id = ?", (idea_id,)
+    ).fetchone()
+    if emb_row:
+        target_vec = np.frombuffer(emb_row["vector"], dtype=np.float32)
+        all_embs = db.all_idea_embeddings()
+        # Compute cosine similarities
+        scores = []
+        for other_id, other_vec in all_embs.items():
+            if other_id == idea_id:
+                continue
+            cos_sim = float(np.dot(target_vec, other_vec) / (
+                np.linalg.norm(target_vec) * np.linalg.norm(other_vec) + 1e-9))
+            scores.append((other_id, cos_sim))
+        scores.sort(key=lambda x: x[1], reverse=True)
+        top_5 = scores[:5]
+
+        # Fetch idea details for top 5
+        if top_5:
+            ids = [s[0] for s in top_5]
+            sim_map = {s[0]: s[1] for s in top_5}
+            placeholders = ",".join("?" * len(ids))
+            sim_rows = db.conn.execute(
+                f"SELECT id, title, idea_type, draft_name FROM ideas WHERE id IN ({placeholders})",
+                ids,
+            ).fetchall()
+            sim_dict = {r["id"]: r for r in sim_rows}
+            for sid, score in top_5:
+                sr = sim_dict.get(sid)
+                if sr:
+                    similar.append({
+                        "id": sr["id"],
+                        "title": sr["title"],
+                        "type": sr["idea_type"],
+                        "draft_name": sr["draft_name"],
+                        "similarity": round(score, 3),
+                    })
+
+    idea["similar"] = similar
+    return idea
+
+
 def get_timeline_data(db: Database) -> TimelineData:
     """Return monthly counts by category for timeline chart."""
     pairs = db.drafts_with_ratings(limit=1000)