[hln33]

Issue #, if available

Reverts #810 (commit db2ea4330b829198b57e8632d785981f99e91144). This restores the telemetry-exporter stabilization from #788.

Description of changes

Commit 1 is git revert db2ea433. It restores exporter version 1.0.0, the Development Status :: 5 - Production/Stable classifier, the telemetry-exporter-v1.0.0 tag example in RELEASING.md, and the two stable-release tests in tests/test_package_metadata.py. It removes the Experimental note from the exporter README. The revert conflicted on one line: __about__.py had moved from 0.0.1 to 0.0.2 in 7f2af31b after #810. The conflict is resolved to 1.0.0. Every other hunk is the exact inverse of #810.

Commit 2 moves the exporter pin in packages/aws-durable-execution-sdk-python-conformance-tests-telemetry/pyproject.toml from ==0.0.2 to ==1.0.0. That package pins the exporter exactly, so the pin moves with the version (same rule 7f2af31b followed).

Exporter runtime behavior and the core plugin interface are unchanged. No package publication, release tag, merge, or auto-merge is requested.

Design

• Design note (rule 3): not needed. This is a revert of version metadata and documentation; no public API changes.
• Simpler options considered (rule 2): no logic is added. A plain git revert alone was considered and rejected because it leaves the conformance package pinned to ==0.0.2, which the workspace test env cannot satisfy once the exporter is 1.0.0.
• Root cause (rule 5): not a bug fix.
• Net change: +24/-8 source lines across 6 files; 0 public API members changed.

Assumptions

• The telemetry conformance CI build installs the exporter from this repository at the PR commit, not from PyPI, so the ==1.0.0 pin does not require a published release. Evidence: packages/aws-durable-execution-sdk-python-conformance-tests-telemetry/src/requirements.txt installs aws-durable-execution-sdk-python-telemetry-exporter @ git+...@${PYTHON_SDK_REF}.
• docs/migration-2.x-to-3.0.md names 0.0.2 as the published exporter version. That statement describes PyPI history, so it is left unchanged.

Decisions not covered by the issue or the design note

• Resolved the __about__.py conflict to 1.0.0 rather than 0.0.2, because the purpose of the PR is to restore the stable version.
• Moved the conformance pin in a separate commit so the revert commit stays a pure inverse of chore: revert telemetry stabilization #810.

Checklist

• Every new branch has a test that fails without it (rule 4) — no new logic; the restored tests test_public_version_matches_installed_stable_release and test_package_is_marked_production_stable fail on current main and pass here.
• For a bug fix: the first commit adds a test that fails on the current code (rule 5) — not a bug fix.
• Every edited test expectation is listed in the description, with the reason (rule 5) — the two restored tests above; no other test changed.

Testing

Validated locally with Hatch 1.18.1 on top of main at 9d704ff6.

• hatch fmt --check: passed; 781 files already formatted.
• hatch run types:check: no issues in all packages.
• hatch run test:all on the exporter and conformance-telemetry tests: 300 passed. The test env resolved the workspace exporter at 1.0.0.
• hatch build -t wheel in the exporter package: Version: 1.0.0, Development Status :: 5 - Production/Stable.

By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.

[github-actions]

Codex AI review

No actionable findings in the six-file diff, including against CONTRIBUTING.md rules 1–9. Residual test risk: the restored metadata tests and package build were not executed under the review constraints.

Reviewed commit b6eac814b029569b49cc0b7176b2e9132f8102df. Workflow run