Agrega script_path + script_bucket para subir scripts PySpark locales
a S3 automáticamente antes de crear Glue Jobs.

Features (item 37):
- Storage::S3#upload_file y Storage::Local#upload_file
- GlueRunner.upload_script(local_path:, bucket:, folder:, filename:)
- create_job y ensure_job aceptan script_path: + script_bucket:
  para upload automático con integracion en diffing

Docs:
- docs/glue-jobs-lifecycle.md: seccion upload local + IAM + concurrencia
- docs/glue_pyspark_example.py: ejemplo script_path
- skill/SKILL.md: FAQ + version bumped

Tests: 185 specs, 0 failures, 0 RuboCop offenses