Implement case-preserving provider ID algorithm v3

[InfoSecHack]

Summary

• Bump ID_ALGORITHM to sha256_null_separated_v3_case_sensitive_provider_ids.
• Preserve case for provider-owned identity fields in node_id and edge_id while keeping structural fields normalized.
• Keep constraint_id on the legacy lowercase canonicalization path pending separate review.
• Convert case-sensitive ARN collision tests from xfail evidence to passing regression coverage.
• Refresh directly affected scenario metadata, golden outputs, and live-binding/demo fixture IDs caused by v3 node/edge ID churn.

ID behavior

• node_id: normalizes provider and node_type; preserves provider_id case.
• edge_id: normalizes edge_type and region; preserves src_provider_id, dst_provider_id, and features_digest.
• constraint_id: unchanged legacy canonical_id behavior.

Validation

• python -m pytest -q tests/identity/test_case_sensitive_id_collision.py: 6 passed
• python -m pytest -q tests/test_deterministic_ids.py tests/test_scenario_output.py tests/test_findings_json.py: 85 passed
• Focused affected golden/live-binding/replay tests: 168 passed
• ./scripts/check.sh: passed
• ./scripts/test_fast.sh: 1996 passed
• git diff --check: passed
• Account/ARN hygiene scans: clean
• Terraform/raw artifact scan: clean