Large HTML documents cannot be partitioned using the partition_html function.

[ratzrattillo]

partitioning large html documents leads to an empty result. This is due to the missing huge_tree option on HTML parser generation in:

https://github.com/Unstructured-IO/unstructured/blob/main/unstructured/partition/html/parser.py#L929

Fix:
Include the huge_tree option to solve this: etree.HTMLParser(remove_comments=True, huge_tree=True)

[CrepuscularIRIS]

Hi — I'd like to work on this.

I reproduced the empty result locally: deeply nested HTML (depth >= 256) passes through etree.HTMLParser(remove_comments=True) in unstructured/partition/html/parser.py and etree.fromstring silently drops the subtree, so partition_html returns zero elements. Adding huge_tree=True to the parser makes <p>-inside-deep-<div> nesting round-trip correctly.

Proposed approach: set huge_tree=True on the module-level html_parser, add a regression test using a deeply-nested fixture, and a CHANGELOG entry — as outlined in the issue. Would a PR in that shape be welcome?