fix: CompactionHook token over-count when no assistant message is present

[RahilOp]

Fixes #12580\n\nWhen context_tokens > 0 but the conversation has no assistant message, _estimated_context_tokens was counting the entire message list on top of the provider's reported usage. This change short-circuits and returns context_tokens when there is no assistant message, matching the behavior described in the issue.\n\nChanges:\n- Updated _estimated_context_tokens in haystack/hooks/compaction/utils.py.\n- Added regression tests in test/hooks/compaction/test_utils.py.\n- Added release note.

[vercel]

@RahilOp is attempting to deploy a commit to the deepset Team on Vercel.

A member of the Team first needs to authorize it.

[CLAassistant]

Thank you for your submission! We really appreciate it. Like many open source projects, we ask that you sign our Contributor License Agreement before we can accept your contribution.
_{You have signed the CLA already but the status is still pending? Let us recheck it.}

[HaystackBot]

Hi @RahilOp, thanks a lot for your contribution! 🙏

We noticed that the Contributor License Agreement (CLA) check (license/cla) hasn't passed yet, so we've temporarily moved this PR to draft and paused the review assignment.

To get your PR reviewed, please sign the CLA via the link in the license/cla check below (or in the CLA bot comment). As soon as the check turns green, this PR will automatically be marked ready for review again and a reviewer will be re-assigned.

[sjrl]

hey @RahilOp thanks for opening the PR! we opted to go for a slightly different fix in this PR #12594 so I'll be closing this one