feat: Add column_order parameter to Tabulator widget

[SuMayaBee]

Description

Adds column_order parameter to Tabulator widget to control column visibility and display order, equivalent to Streamlit's data_editor column_order parameter.

Changes

• Add column_order parameter to filter and reorder displayed columns
• Override _get_columns() to implement column filtering/ordering logic
• Add comprehensive tests

Example

import pandas as pd
import panel as pn

df = pd.DataFrame({
    'col1': [1, 2, 3],
    'col2': [4, 5, 6],
    'col3': [7, 8, 9],
    'col4': [10, 11, 12],
})

# Show only col3 and col1, in that order
tab = pn.widgets.Tabulator(df, column_order=['col3', 'col1'])

The table will display columns in the order: index, col3, col1 (col2 and col4 are hidden).

Behavior

• column_order=None (default): All columns shown in original order
• column_order=[...]: Only listed columns shown in specified order
• Index columns always appear first and cannot be reordered/hidden
• Invalid column names are silently ignored

Fixes #8431

AI Disclosure: Claude Haiku 4.5 was used for exploring the codebase during the development of this PR.

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 86.00%. Comparing base (49872f3) to head (6dba194).
⚠️ Report is 48 commits behind head on main.

Additional details and impacted files

@@            Coverage Diff             @@
##             main    #8434      +/-   ##
==========================================
- Coverage   86.12%   86.00%   -0.13%     
==========================================
  Files         349      349              
  Lines       54971    55013      +42     
==========================================
- Hits        47343    47313      -30     
- Misses       7628     7700      +72     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.

[SuMayaBee]

Hi @philippjfr, I’ve added a small visual demo to verify the new column_order behavior.

pn.widgets.Tabulator(df)                              # default: all columns in original order
pn.widgets.Tabulator(df, column_order=["col3", "col1"])  # only col3 and col1 shown, in that order; others hidden

In the attached screenshot, both sides use the same input DataFrame:

• Left: default Tabulator (column_order=None) → all columns are shown in the original order.
• Right: column_order=['col3', 'col1'] → only col3 and col1 are displayed, in that order, with the index preserved and col2/col4 hidden.

This matches the behavior described in #8431 and Streamlit’s data_editor.column_order parameter.
Whenever you have time, I’d really appreciate a review or any feedback on the approach. Thanks!