Fix misleading order in readme

[JulianJvn]

The BM25 score is in range $(-\infty, 0]$, where more negative scores indicate more relevant documents. However, the readme says higher scores are more relevant. A score near 0 is higher than a more negative score, so this is wrong. It cost us quite some time and caused a lot of confusion to figure this out.

[github-actions]

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 can sign the CLA by just posting a Pull Request Comment same as the below format.

---

I have read the CLA Document and I hereby sign the CLA

---

_{You can retrigger this bot by commenting recheck in this Pull Request. Posted by the CLA Assistant Lite bot.}

[JulianJvn]

I don't have and don't want to get permission from my employer to sign that CLA just for this single line change. If someone else who has/can sign it could recreate my fix from scratch, it would be much appreciated since this will save confusion for others as well.

[VoVAllen]

Thanks for the correction! You're right, we change the BM25 score sign here so the description here is kind of misleading. Have you met other problems with VectorChord-bm25? Would love to hear any feedbacks!

[VoVAllen]

Thanks for your help! I've updated it in #86