Fix paper_index paths and add GOLD, IW-OPD, and AsyncDistillation telemetry

[YeonwooSung]

What does this PR do?

Small docs and telemetry hygiene on current main:

• paper_index.md pointed Online DPO at the nonexistent experimental.odpo module. The real package is experimental.online_dpo.
• The LD-DPO sentence was truncated ("recovers standard α reduces verbosity"). It now says α=1 recovers standard DPO and smaller α reduces verbosity.
• "TRL also provide" → "provides".
• GOLD only appeared as a drive-by snippet under the Thinking Machines OPD blog. It now has its own subsection, using the project page from GOLDTrainer._paper (no papers id in that citation).
• IW-OPD already has a paper_index entry and tests, but no docs page. Adds docs/source/iw_opd_trainer.md and an Experimental toctree row.
• AsyncDistillationTrainer was missing from _TELEMETRY_TRAINERS, so it reported as "other".

No trainer behavior changes except the one-line telemetry allowlist.

Fixes #7048

Before submitting

• This PR fixes a typo or improves the docs (you can dismiss the other checks if that's the case).
• Did you read the contributor guideline, Pull Request section?
• Was this discussed/approved via a GitHub issue? Please add a link to it if that's the case.
• Did you make sure to update the documentation with your changes?
• Did you write any new necessary tests?

AI writing disclosure

We welcome the use of AI tools to help with contributions. For transparency and to help us improve our review process, please indicate the level of AI involvement in this PR.

• No AI usage: the PR was written entirely by a human.
• AI-assisted: some parts were suggested or improved by AI, but the PR was written and reviewed by a human.
• AI-generated: the PR was mostly or fully generated by an AI tool.

Who can review?

Anyone in the community is free to review the PR once the tests have passed.

---

Note

Low Risk
Changes are limited to documentation and a one-line telemetry allowlist entry; no training, auth, or data-handling behavior is modified.

Overview
This PR is mostly documentation and telemetry hygiene—no trainer training logic changes beyond the telemetry allowlist.

Paper index fixes broken or unclear references: Online DPO now points at experimental.online_dpo instead of the nonexistent experimental.odpo, the LD-DPO blurb is completed (α=1 recovers standard DPO), and a small grammar fix (“TRL also provides…”). The distillation section gains a dedicated GOLD subsection with project context and a GOLDConfig example that highlights use_uld_loss=True for cross-tokenizer distillation.

IW-OPD gets a first-class experimental docs page (iw_opd_trainer.md) with usage, IWOPDConfig/IWOPDTrainer autodoc, and a matching Experimental toctree entry.

Telemetry: AsyncDistillationTrainer is added to _TELEMETRY_TRAINERS in base_trainer.py so usage is reported under the correct trainer name instead of "other".

^{Reviewed by Cursor Bugbot for commit 9e8e016. Bugbot is set up for automated code reviews on this repo. Configure here.}