convert_tables: don't crash on tables with no rows (#63 case 1)

[eeshsaxena]

Picks off the unambiguous part of #63: an empty table crashes.

import html_to_json
html_to_json.convert_tables('<table></table>')
html_to_json.convert_tables('<table><tbody></tbody></table>')
# IndexError: list index out of range

_process_table reads find_all('tr')[0] before checking that the table has any rows. An empty <tbody> is ordinary CMS output, so this isn't only an adversarial-input concern.

Added an early return of the empty result when there are no rows. convert_tables already drops empty results, so a rowless table is simply omitted (including when it sits next to a real table). Tables with rows are untouched.

I deliberately left the ragged/class-B cases from #63 out of this PR since, as you noted there, they need a semantics decision and case 2 overlaps with #64.

Added two tests in test_html_tables_misc.py (empty table alone, and empty table next to a real one). The empty-table case raises IndexError on main and passes with this change.

[eeshsaxena]

Hi! Gentle nudge on this one whenever you have some bandwidth. It's a small, self-contained fix (convert_tables: don't crash on tables with no rows (#63 case 1)), and it's currently mergeable with no conflicts. No urgency at all, and I'm happy to make any changes you'd like. Thanks for maintaining html-to-json!