use_columns silently skips empty columns at the beginning

[jsarbach]

Given the following spreadsheet with an empty first column:

Loading the sheet with e.g. use_columns='B:C' (or use_columns=[1, 2]) actually selects columns C and D:

read_excel('sheet.xlsx').load_sheet(0, use_columns='B:C').to_polars()
shape: (3, 2)
┌────────┬───────┐
│ Second ┆ Third │
│ ---    ┆ ---   │
│ f64    ┆ f64   │
╞════════╪═══════╡
│ 4.0    ┆ 7.0   │
│ 5.0    ┆ 8.0   │
│ 6.0    ┆ 9.0   │
└────────┴───────┘

And raises ColumnNotFoundError with use_columns='B:D' or use_columns=[1, 2, 3]:

read_excel('sheet.xlsx').load_sheet(0, use_columns='B:D').to_polars()
_fastexcel.ColumnNotFoundError: column at index 3 not found
Context:
    0: available columns are: [ColumnInfo { name: "First", index: 0, dtype: Float, column_name_from: LookedUp, dtype_from: Guessed }, ColumnInfo { name: "Second", index: 1, dtype: Float, column_name_from: LookedUp, dtype_from: Guessed }, ColumnInfo { name: "Third", index: 2, dtype: Float, column_name_from: LookedUp, dtype_from: Guessed }]

[PrettyWood]

Yes the engine behind fastexcel calamine skips empty columns by default.
I plan to add an option on calamine directly like I did for skipping empty rows