Convert numpy integer and float scalars to Python numbers on insert

[TanbirRamim]

Fixes #876.

Raw numpy scalars like np.int64(5) were passed straight to sqlite3, which stores them via the buffer protocol as BLOBs (b'\x05\x00...'). The column types were already right, since COLUMN_TYPE_MAPPING knows the numpy types, but the values were not. jsonify_if_needed() now calls .item() on np.integer / np.floating values when numpy is installed.

I left out other numpy types (bool_, datetime64, ...) because they aren't in the type mapping yet and .item() isn't always what you'd want for them.

Tests: added test_insert_raw_numpy_scalars, which uses pytest.importorskip("numpy"), so it runs in the numpy: 1 CI legs without needing pandas. It fails on main and passes with the fix. The full suite passes with and without numpy installed. black, flake8, mypy, pyright and ty check are clean. This doesn't touch the test that #875 rewrites.

---

📚 Documentation preview 📚: https://sqlite-utils--877.org.readthedocs.build/en/877/

[Copilot]

Copilot was unable to review this pull request because the user who requested the review has reached their quota limit.