Add lifecycle hook telemetry for before_all/after_all transparency

[christso]

Problem

before_all failures are visible, but success-path transparency is limited. It's hard to quickly confirm exactly what command ran, where it ran, and timing details unless custom logs are added in scripts.

Request

Add built-in lifecycle hook telemetry for before_all/after_all/before_each/after_each in eval results/CLI output.

Suggested fields:

• hook name
• resolved command array
• resolved cwd
• start/end timestamps
• duration
• exit code
• stdout/stderr snippets (or links)

Acceptance

• Hook execution details are shown by default in eval output (at least once per run).
• Success and failure paths both include telemetry.
• No custom script logging required to audit hook execution.

Why

Improves transparency/debuggability and reduces the need for extra boilerplate logs in setup scripts.