test(backend): add unit tests for PDF chunker table parsing

[nancysangani]

🔗 Related Issue

Closes #451

---

📝 What does this PR do?

Extends backend/tests/test_chunker.py with 11 new unit tests that verify
PDF table parsing across all three extraction paths and edge cases. All tests
use in-memory fakes and monkeypatching — no real PDF files are committed and
no new dependencies are introduced.

_table_to_markdown edge cases:

• Empty row list returns empty string
• All-blank cells return empty string
• Single header row renders correctly with separator
• Ragged rows (unequal column counts) are padded to uniform width
• Whitespace in cells (tabs, newlines) is normalised to single spaces
• Pipe characters in cells are escaped as \|
• Separator row always uses ---

pdfplumber path:

• Multi-page PDF produces separate table chunks with correct page numbers
• Table whose cells are all blank emits no chunk
• table_index increments correctly across multiple tables on one page
• Normalised bbox values are each within [0.0, 1.0]

PyMuPDF fallback path:

• When unstructured and pdfplumber are both absent, PyMuPDF still produces
  chunk_type=text chunks

Image + table interleaving:

• Image chunks are appended after table/text chunks on the same page

chunk_index continuity:

• chunk_index is 0-based and monotonically increasing across the full
  document with no resets or gaps

---

🗂️ Type of Change

• 🧪 Tests

---

🧪 How was this tested?

• pytest backend/tests/test_chunker.py -v — all 11 new tests pass
  alongside the 7 existing tests (18 total, 0 failures)
• No real PDF files used — all fixtures are in-memory fakes
• No new imports outside stdlib + existing project deps

---

✅ Self-Review Checklist

• My branch is based on dev, not main
• I have not added any secrets / API keys
• I have not modified main branch or any HuggingFace deployment config
• My code follows the existing style (no unnecessary formatting changes)
• I have updated relevant docs / comments if needed

[nancysangani]

Hi @param20h, I have opened this PR to fix the issue #451. Please review it when you get a chance. Thanks!

[nancysangani]

@param20h please review this PR, all checks have passed and resolved the conflicts. Thanks!