data_utils.py has a bug in extract_pdf_content

[cynthiajiangatl]

Describe the bug
When there is an empty table in a pdf document, if layout model is used, extract_pdf_content will fail with "list index out of range" error.

Expected behavior
Empty table should be skipped.

Code fix needed
Add try catch and skip the empty table.

for table in form_recognizer_results.tables:
try:
table.spans[0]
except:
continue
table_offset = table.spans[0].offset
table_length = table.spans[0].length
if page_offset <= table_offset and table_offset + table_length < page_offset + page_length:
tables_on_page.append(table)

[vkrd]

Thanks for pointing this out, fixed in #1040