Set up Docker Compose with Postgres 18 and pgvector 0.8.1 extension.

Purpose: Provides the database foundation for storing line items with vector embeddings for semantic search comparison.

Output: Running Postgres database with line_item table ready for data import and embedding storage.

<execution_context> @./.claude/get-shit-done/workflows/execute-plan.md @./.claude/get-shit-done/templates/summary.md </execution_context>

@.planning/PROJECT.md @.planning/ROADMAP.md @.planning/STATE.md @.planning/phases/01-foundation/01-CONTEXT.md @.planning/phases/01-foundation/01-RESEARCH.md Task 1: Create Docker Compose configuration docker-compose.yml Create docker-compose.yml with Postgres 18 + pgvector service: 
services:
  db:
    image: pgvector/pgvector:pg18
    container_name: semantic-search-db
    ports:
      - "5433:5432"
    environment:
      POSTGRES_USER: dev
      POSTGRES_PASSWORD: dev
      POSTGRES_DB: semantic_search
    volumes:
      - ./init.sql:/docker-entrypoint-initdb.d/init.sql
      - pgvector_data:/var/lib/postgresql/18/docker
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U dev -d semantic_search"]
      interval: 5s
      timeout: 5s
      retries: 5

volumes:
  pgvector_data:

Use port 5433 to avoid conflicts with local Postgres installations. Use Postgres 18 data path /var/lib/postgresql/18/docker (changed from /data in Postgres 18). docker compose config (validates YAML syntax) docker-compose.yml exists with valid pgvector/pgvector:pg18 image configuration

Task 2: Create database initialization script init.sql Create init.sql with pgvector extension and line_item table schema: 
-- Enable pgvector extension
CREATE EXTENSION IF NOT EXISTS vector;

-- Line item table matching Orcha's booking_history_item structure
CREATE TABLE line_item (
    id BIGSERIAL PRIMARY KEY,
    -- Original values
    supplier_name TEXT NOT NULL,
    description TEXT NOT NULL,
    debit_account TEXT NOT NULL,
    credit_account TEXT,
    cost_center TEXT,
    net_amount NUMERIC,
    -- Normalized values (for matching)
    supplier_name_normalized TEXT NOT NULL,
    description_normalized TEXT NOT NULL,
    -- Embedding columns (populated in Phase 2, defined now to avoid migrations)
    embedding_google vector(768),      -- text-multilingual-embedding-002
    embedding_jina vector(1024),       -- jina-embeddings-v3
    embedding_minilm vector(384),      -- all-MiniLM-L6-v2
    -- Metadata
    created_at TIMESTAMPTZ NOT NULL DEFAULT now()
);

-- Note: HNSW indexes created in Phase 2 after data load for performance

Per user decision: Include nullable embedding columns from start to avoid migrations. Vector dimensions per research: Google=768, Jina=1024, MiniLM=384. Check SQL syntax: docker compose up -d && sleep 3 && docker compose exec db psql -U dev -d semantic_search -c "\d line_item" init.sql creates vector extension and line_item table with all required columns including embedding vectors

Task 3: Verify database setup Start the database and verify everything is working:
  1. Start container: docker compose up -d
  2. Wait for healthy: docker compose ps (should show healthy)
  3. Verify pgvector: docker compose exec db psql -U dev -d semantic_search -c "SELECT extversion FROM pg_extension WHERE extname = 'vector';"
  4. Verify table: docker compose exec db psql -U dev -d semantic_search -c "\d line_item"
  5. Verify vector columns: docker compose exec db psql -U dev -d semantic_search -c "SELECT column_name, data_type FROM information_schema.columns WHERE table_name = 'line_item' AND data_type = 'USER-DEFINED';"

Stop container after verification (will be started again in Plan 02): docker compose down All verification commands pass without errors Database container starts, pgvector extension is enabled, line_item table has correct schema with vector columns

1. `docker compose config` validates configuration 2. Container starts and becomes healthy within 30 seconds 3. pgvector extension version is returned from pg_extension query 4. line_item table shows all columns including embedding_google, embedding_jina, embedding_minilm 5. Vector columns are type USER-DEFINED (pgvector's vector type)

<success_criteria>

After completion, create `.planning/phases/01-foundation/01-01-SUMMARY.md`