feat(flagd): Fix in-process provider to use FLAGD_SYNC_PORT for configuration

[Copilot]

• [x] Add FLAGD_SYNC_PORT environment variable constant
• [x] Update updateFromEnvVar() to read FLAGD_SYNC_PORT first, then fallback to FLAGD_PORT for backwards compatibility
• [x] Update README documentation to reflect the environment variable changes
• [x] Add unit tests for the new FLAGD_SYNC_PORT functionality
• [x] Run existing tests to ensure no regressions
• [x] Run code review and security checks
• [x] Revert unintentional go.mod and go.sum changes

Original prompt

---

This section details on the original issue you should resolve

<issue_title>[flagd] Fix in-process provider to use FLAGD_SYNC_PORT (not FLAGD_PORT) for port configuration (backwards compatibility)</issue_title> <issue_description>## Background

Recent updates to flagd and related tests in flagd-testbed clarified the use of environment variables for ports:

• FLAGD_PORT is for remote/server mode.
• FLAGD_SYNC_PORT should be used for in-process/sync providers.

Currently, the in-process provider uses the wrong environment variable (FLAGD_PORT).

Required Changes

• Fix provider configuration logic to use FLAGD_SYNC_PORT for in-process provider setups, and fall back to FLAGD_PORT for backwards compatibility.
• Update any relevant documentation and configuration references.
• Ensure test coverage, ideally using scenarios matching those in flagd-testbed (with @legacy-sync-port tag for backwards compatibility coverage).

References

• Parent: https://github.com/open-feature/flagd/issues/1573
• Testbed: https://github.com/open-feature/flagd-testbed/issues/266</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes open-feature/go-sdk-contrib#788

---

💡 You can make Copilot smarter by setting up custom instructions, customizing its development environment and configuring Model Context Protocol (MCP) servers. Learn more Copilot coding agent tips in the docs.

[aepfli]

@copilot please revert all your go.mod and go.sum updates, there is no need to do this within this pullrequest

[Copilot]

@copilot please revert all your go.mod and go.sum updates, there is no need to do this within this pullrequest

Reverted all go.mod and go.sum changes in 5f67528.

[aepfli]

/gemini review

[erka]

@aepfli Could you please change the PR and issue title? feat(..) Fix... doesn't sound right and doesn't explain the change very well.

[aepfli]

I will update the repo, we actually do have gherkin tests in the testbed, and this test by copilot is not needed.

[aepfli]

@erka i removed the test generated by copilot - and added the gherkin testsuite tests for this kind of action - which was good because it catched issues in the gherkin files, and in the implementation (as we did not know which resolver we are in, at the previous implementation)