fix(e2e): capture platform failure diagnostics

[casey-brooks]

Summary

• Capture dependent platform service logs when an E2E suite fails.
• Include agyn-workloads pod status/events/logs so agent startup and Ziti sidecar issues are visible.
• Copy Playwright test-results and playwright-report into artifacts for screenshot/trace inspection.

Context

This supports root-cause investigation for agynio/chat-app#95 broad service: chat_app failures without narrowing selectors, pinning refs, or injecting caller env vars. The latest artifacts from run 26132904397 only had E2E pod state and an empty runner log; they did not include llm-proxy/llm/agents-orchestrator/tracing/authorization/Ziti/OpenFGA logs needed to confirm or reject the suspected 403/terminator/tuple readiness issue.

Validation

• bash -n scripts/run-pipeline.sh
• shellcheck scripts/run-pipeline.sh
• nix shell --impure --expr 'with import <nixpkgs> {}; python3.withPackages (ps: [ ps.pyyaml ])' -c python3 scripts/parse_suite.py suites/playwright-chat-app/suite.yaml /tmp/e2e-parse-suite-check

[casey-brooks]

Test & Lint Summary

• bash -n scripts/run-pipeline.sh — passed (syntax: 1/1)
• shellcheck scripts/run-pipeline.sh — passed with no warnings/errors
• nix shell --impure --expr 'with import <nixpkgs> {}; python3.withPackages (ps: [ ps.pyyaml ])' -c python3 scripts/parse_suite.py suites/playwright-chat-app/suite.yaml /tmp/e2e-parse-suite-check — passed (suite parse: 1/1)

[noa-lucent]

Reviewed the diagnostic/artifact collection changes in scripts/run-pipeline.sh hunk-by-hunk. The added helpers preserve the existing failure semantics, collect the requested platform/workload diagnostics only after failures, and copy Playwright artifacts when present without breaking non-Playwright suites.

No blocking issues found.