Migrate to modern bs4 interface (#4025)

## PR Summary
This small PR fixes the bs4 deprecation warnings which you can find in
the [CI
logs](https://github.com/Unstructured-IO/unstructured/actions/runs/15491657572/job/43729960936#step:3:2615):
```python
/app/unstructured/metrics/table/table_extraction.py:53: DeprecationWarning: Call to deprecated method findAll. (Replaced by find_all) -- Deprecated since version 4.0.0.
/app/unstructured/metrics/table/table_extraction.py:57: DeprecationWarning: Call to deprecated method findAll. (Replaced by find_all) -- Deprecated since version 4.0.0.
```

---------

Signed-off-by: Emmanuel Ferdman <emmanuelferdman@gmail.com>

CHANGELOG.md

@@ -1,4 +1,4 @@
-## 0.17.11-dev2
+## 0.17.11-dev3
 
 ### Enhancements
 
@@ -8,6 +8,7 @@
 - Fix type error when `result_file_type` is expected to be a `FileType` but is `None`
 - Fix chunking for elements with None text that has AttributeError 'NoneType' object has no attribute 'strip'.
 - Invalid elements IDs are not visible in VLM output. Parent-child hierarchy is now retrieved based on unstructured element ID, instead of id injected into HTML code of element.
+- Fix bs4 deprecation warnings by updating `findAll()` with `find_all()`.
 
 ## 0.17.10
 - Drop Python 3.9 support as it reaches EOL in October 2025

unstructured/__version__.py

@@ -1 +1 @@
-__version__ = "0.17.11-dev2"  # pragma: no cover
+__version__ = "0.17.11-dev3"  # pragma: no cover

unstructured/metrics/table/table_extraction.py

@@ -50,11 +50,11 @@ def html_table_to_deckerd(content: str) -> List[Dict[str, Any]]:
 
     soup = BeautifulSoup(content, "html.parser")
     table = soup.find("table")
-    rows = table.findAll(["tr"])
+    rows = table.find_all(["tr"])
     table_data = []
 
     for i, row in enumerate(rows):
-        cells = row.findAll(["th", "td"])
+        cells = row.find_all(["th", "td"])
         for j, cell_data in enumerate(cells):
             cell = {
                 "y": i,