Fix database cleanup and technical inaccuracies per PR feedback

- Preserve schema_migrations in parallel test cleanup (prevents migration tracking breakage)
- Add missing pgvector gem to Gemfile examples in both blog post and SEO doc
- Correct IVFFlat description from "exact search" to "approximate ANN index"

🤖 Generated with [Claude Code](https://claude.com/claude-code)

Co-Authored-By: Claude <[email protected]>

Author: pftg

content/blog/2025/pgvector-rails-tutorial-production-semantic-search.md

@@ -37,7 +37,7 @@ pgvector is a PostgreSQL extension that enables vector similarity search directl
 **Core Features**:
 - **Vector Storage**: Efficient storage for high-dimensional vectors (1-16,000 dimensions)
 - **Distance Functions**: Cosine similarity, L2 distance, inner product
-- **Index Types**: HNSW (fast approximate search) and IVFFlat (scalable exact search)
+- **Index Types**: HNSW (fast approximate ANN search) and IVFFlat (scalable approximate ANN index)
 - **Rails Integration**: neighbor gem provides ActiveRecord-friendly interface
 
 **Production Readiness**:

content/blog/2025/tdd-workflow-automation-rails-teams.md

@@ -380,9 +380,14 @@ class ActiveSupport::TestCase
     # Configure ActiveRecord for parallel execution
     parallelize(workers: :number_of_processors) if respond_to?(:parallelize)
 
-    # Ensure clean database state
+    # Clean database state (preserve schema_migrations for Rails migration tracking)
     parallelize_setup do |worker|
-      ActiveRecord::Base.connection.execute("TRUNCATE TABLE schema_migrations RESTART IDENTITY CASCADE")
+      # Rails creates isolated test databases per worker via parallel:create parallel:prepare
+      # Only truncate data tables, exclude schema_migrations to preserve migration state
+      tables_to_truncate = ActiveRecord::Base.connection.tables - ['schema_migrations']
+      tables_to_truncate.each do |table|
+        ActiveRecord::Base.connection.execute("TRUNCATE TABLE #{table} RESTART IDENTITY CASCADE")
+      end
     end
 
     parallelize_teardown do |worker|

docs/projects/2510-seo-content-strategy/scheduled-posts/priority-2-pgvector-rails.md

@@ -202,6 +202,7 @@ end
 #### Step 3: Install neighbor Gem
 ```ruby
 # Gemfile
+gem 'pgvector'     # PostgreSQL vector extension support
 gem 'neighbor'
 gem 'ruby-openai' # For generating embeddings
 ```