docs(open_feature): add LLM coding guide for FFE contributors

[TonyCTHsu]

What does this PR do?
Adds a scoped AGENTS.md coding guide under lib/datadog/open_feature/ for contributors and their AI coding tools working in that subtree.

Motivation:
The OpenFeature area receives LLM-assisted contributions from a team without a dedicated Ruby engineer. Review time is spent correcting conventions rather than evaluating design and correctness. This guide codifies the conventions upfront so generated code arrives closer to the bar we hold everywhere else in the tracer.

Change log entry
None.

Additional Notes:
The file is scoped to lib/datadog/open_feature/ so it loads lazily and adds zero context overhead for maintainers working elsewhere. Each rule is grounded in a real finding from this codebase; the guide is intended to grow as new patterns emerge.

How to test the change?
Documentation only — no code changes.

[datadog-prod-us1-5]

🎉 All green!

🧪 All tests passed
❄️ No new flaky tests detected

🔄 Datadog auto-retried 1 job - 0 passed on retry

🎯 Code Coverage (details)
• Patch Coverage: 100.00%
• Overall Coverage: 90.02% (-0.00%)

_{This comment will be updated automatically if new data arrives.
🔗 Commit SHA: 709537d | Docs | Datadog PR Page | Give us feedback!}

[y9v]

awesome guide! Thank you, this will save us all a lot of time

[pr-commenter]

Benchmarks

Benchmark execution time: 2026-06-30 12:17:11

Comparing candidate commit 4dee0fd in PR branch tony.hsu/open-feature-agents-guide with baseline commit 4550c8e in branch master.

Found 0 performance improvements and 0 performance regressions! Performance is the same for 48 metrics, 1 unstable metrics.

Explanation

This is an A/B test comparing a candidate commit's performance against that of a baseline commit. Performance changes are noted in the tables below as:

• 🟩 = significantly better candidate vs. baseline
• 🟥 = significantly worse candidate vs. baseline

We compute a confidence interval (CI) over the relative difference of means between metrics from the candidate and baseline commits, considering the baseline as the reference.

If the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD), the change is considered significant.

Feel free to reach out to #apm-benchmarking-platform on Slack if you have any questions.

More details about the CI and significant changes

You can imagine this CI as a range of values that is likely to contain the true difference of means between the candidate and baseline commits.

CIs of the difference of means are often centered around 0%, because often changes are not that big:

---------------------------------(------|---^--------)-------------------------------->
                              -0.6%    0%  0.3%     +1.2%
                                 |          |        |
         lower bound of the CI --'          |        |
sample mean (center of the CI) -------------'        |
         upper bound of the CI ----------------------'

As described above, a change is considered significant if the CI is entirely outside the configured SIGNIFICANT_IMPACT_THRESHOLD (or the deprecated UNCONFIDENCE_THRESHOLD).

For instance, for an execution time metric, this confidence interval indicates a significantly worse performance:

----------------------------------------|---------|---(---------^---------)---------->
                                       0%        1%  1.3%      2.2%      3.1%
                                                  |   |         |         |
       significant impact threshold --------------'   |         |         |
                      lower bound of CI --------------'         |         |
       sample mean (center of the CI) --------------------------'         |
                      upper bound of CI ----------------------------------'

[anmarchenko]

I think this is a very good guidance! we can adopt it for the whole library as well

[leoromanovsky]

Approving, as I appreciate the direction this is going. I haven't tested it out yet, but looking forward to my next contribution and seeing its effect.

[TonyCTHsu]

Opened #5989 to address the 4 outstanding review threads on this PR (blank line after the guard clause, Zeitwerk file-structure section, root AGENTS.md pointer, and trimming the guide toward ~200 lines with a "Must follow" checklist up top).

Given #5968 already has 5 approvals and the open threads are non-blocking for this docs-only change, planning to merge this PR once #5989 is up for review — the follow-up PR is where the review feedback gets addressed and re-reviewed.