feat(ml): add database schema, config parser, and DVC setup

- Initialize DVC with local storage backend (task 1.6)
- Create PostgreSQL schema for training_runs table (task 1.7)
- Add SQLAlchemy database connection setup (task 1.8)
- Create Pydantic config models for pipeline.yaml (task 2.1)
- Add migration runner for database setup
- Fix pyproject.toml package discovery config

services/ml/migrations/001_create_training_runs.sql

@@ -0,0 +1,27 @@
+-- Create training_runs table for tracking ML training runs
+CREATE TABLE IF NOT EXISTS training_runs (
+    id SERIAL PRIMARY KEY,
+    run_id VARCHAR(255) NOT NULL UNIQUE,
+    model_type VARCHAR(100) NOT NULL,
+    experiment_name VARCHAR(255) NOT NULL,
+    pipeline_config_hash VARCHAR(64) NOT NULL,
+    dataset_version VARCHAR(100),
+    metrics_summary JSONB,
+    status VARCHAR(50) NOT NULL DEFAULT 'running',
+    created_at TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,
+    completed_at TIMESTAMP WITH TIME ZONE,
+    
+    CONSTRAINT valid_status CHECK (status IN ('running', 'completed', 'failed', 'cancelled'))
+);
+
+-- Create index on run_id for faster lookups
+CREATE INDEX idx_training_runs_run_id ON training_runs(run_id);
+
+-- Create index on experiment_name for filtering by experiment
+CREATE INDEX idx_training_runs_experiment ON training_runs(experiment_name);
+
+-- Create index on status for filtering active runs
+CREATE INDEX idx_training_runs_status ON training_runs(status);
+
+-- Create index on created_at for chronological queries
+CREATE INDEX idx_training_runs_created_at ON training_runs(created_at DESC);