feat: overhaul bundled template catalog to align with new engine features

[mohossam01]

What this PR does

Ships six bundled domain templates — banking, health, hr, marketing, retail, saas — each producing a realistic multi-table dataset for its domain in one call, with no config editing required. The catalog overhaul threads the engine's newer surfaces — heteroscedastic and heavy-tailed noise, per-metric treatment cohorts, holdout splits, parent/child fact grain — into every template.

Files Modified

CHANGELOG.md
CONTRIBUTING.md
docs/site/api-reference.md
docs/site/column-types.md
docs/site/config-reference.md
docs/site/cookbook/data-engineering.md
docs/site/cookbook/data-science.md
docs/site/feature-reference.md
docs/site/tutorial-notebooks/getting_started.ipynb
docs/site/user-guide/cdc-facts.md
docs/site/user-guide/experiments-and-cohorts.md
docs/site/user-guide/geo-hierarchy.md
docs/site/user-guide/metrics-and-connections.md
docs/site/user-guide/multi-source.md
docs/site/user-guide/narrative-source.md
plotsim/configs/templates/banking.yaml
plotsim/configs/templates/banking_template.py
plotsim/configs/templates/education_template.py
plotsim/configs/templates/education_template.yaml
plotsim/configs/templates/health.yaml
plotsim/configs/templates/health_template.py
plotsim/configs/templates/hr.yaml
plotsim/configs/templates/hr_template.py
plotsim/configs/templates/hr_template.yaml
plotsim/configs/templates/marketing.yaml
plotsim/configs/templates/marketing_template.py
plotsim/configs/templates/marketing_template.yaml
plotsim/configs/templates/retail.yaml
plotsim/configs/templates/retail_template.py
plotsim/configs/templates/saas.yaml
plotsim/configs/templates/saas_template.py
tests/configs/init.py
tests/configs/ab_trial.py
tests/configs/ab_trial.yaml
tests/configs/bare_minimum.py
tests/configs/bare_minimum.yaml
tests/configs/cdc_demo.py
tests/configs/cdc_demo.yaml
tests/configs/crm_billing_overlap.py
tests/configs/crm_billing_overlap.yaml
tests/configs/geo_retail.py
tests/configs/geo_retail.yaml
tests/configs/lakehouse.py
tests/configs/lakehouse.yaml
tests/configs/latency_skew.py
tests/configs/latency_skew.yaml
tests/configs/narrative_reviews.py
tests/configs/narrative_reviews.yaml
tests/configs/orders_template.py
tests/configs/orders_template.yaml
tests/configs/retail_template.py
tests/configs/retail_template.yaml
tests/configs/saas_template.py
tests/configs/saas_template.yaml
tests/test_builder_input.py
tests/test_builder_integration.py
tests/test_builder_schema.py
tests/test_bypass_observability.py
tests/test_cdc_facts.py
tests/test_cli.py
tests/test_entity_expansion.py
tests/test_fk_target_resolution.py
tests/test_geo_provider.py
tests/test_jsonl_output.py
tests/test_multi_source.py
tests/test_narrative_source.py
tests/test_parent_child_facts.py
tests/test_partitioned_parquet.py
tests/test_pk_prefix.py
tests/test_sql_output.py
tests/test_streaming_parquet.py
tests/test_templates_api.py

How to test

python -m pytest tests/ -q
python -c "import plotsim; assert plotsim.list_templates() == ['banking', 'health', 'hr', 'marketing', 'retail', 'saas']"
for n in banking health hr marketing retail saas; do
  plotsim run "$n" -o "/tmp/plotsim_$n" && plotsim validate "/tmp/plotsim_$n"
done