bug/html parsing incorrectly categorizing text

[bhoppeadoy]

Describe the bug
The html parser is incorrectly categorizing a set of divs as different text types. There are three divs, one right after each other and each div is categorized as UncategorizedText, Title, and NarrativeText. These should all be the same text category and definitely not a title category.

To Reproduce
Parse the attached html file. Examine the elements created for these three div ids SBOS510440, SBOS5102933, SBOS5105314

< div id="SBOS510440" class="textnote">< span class="section-label">1. </ span>V< sub>IN</ sub> = +V< sub>S</ sub> + 500 mV.</ div>

< div id="SBOS5102933" class="textnote">< span class="section-label">2. </ span>TVS: +V< sub>S(max)</ sub> > V< sub>TVSBR (Min)</ sub> > +V_{S</ sub> </ div>}

< div id="SBOS5105314" class="textnote">< span class="section-label">3. </ span>Suggested value is approximately 5 kΩ in overvoltage conditions.</ div>

elements = partition_html(text=html_text)
electrical-overstress-sbos8126148.zip

Expected behavior
Expect these to three divs to be the same text category, but not a title category

Screenshots
Here's the list of elements for the three divs:

Environment Info
OS version: Linux-6.1.83-4.ph5-x86_64-with-glibc2.28
Python version: 3.10.14
unstructured version: 0.15.13
unstructured-inference is not installed
pytesseract is not installed
Torch is not installed
Detectron2 is not installed
PaddleOCR is not installed
Libmagic version: file-5.33
magic file from /etc/magic:/usr/share/misc/magic

[scanny]

Yep, good call. I'll sort this out.