How do Site Reliability Engineers leverage Python to automate repetitive operational toil, build custom telemetry exporters, and orchestrate self-healing incident remediation scripts across distributed architectures, and why is programmatic automation the single most effective way to protect service level objectives and scale infrastructure reliably?