Raise a friendly message when a str is provided to TensorFrame(col_names_dict) instead of a list[str]

[akihironitta]

Repro

import torch
import torch_frame
from torch_frame.data import TensorFrame

feat_dict = {torch_frame.categorical: torch.randint(0, 3, size=(10, 1))}
col_names_dict = {torch_frame.categorical: 'cat_1'}
TensorFrame(feat_dict, col_names_dict)

Before this PR

Traceback (most recent call last):
  File "/home/aki/work/github.com/kumo-ai/kumo/../../akihironitta/gist/test.py", line 7, in <module>
    TensorFrame(feat_dict, col_names_dict)
  File "/home/aki/miniconda3/envs/kumo/lib/python3.10/site-packages/torch_frame/data/tensor_frame.py", line 78, in __init__
    self.validate()
  File "/home/aki/miniconda3/envs/kumo/lib/python3.10/site-packages/torch_frame/data/tensor_frame.py", line 113, in validate
    raise ValueError(
ValueError: The expected number of columns for categorical feature is 5, which does not align with the column dimensionality of feat_dict[categorical] (got 1)

After this PR

Traceback (most recent call last):
  File "/home/aki/work/github.com/pyg-team/pytorch-frame/../../akihironitta/gist/test.py", line 7, in <module>
    TensorFrame(feat_dict, col_names_dict)
  File "/home/aki/work/github.com/pyg-team/pytorch-frame/torch_frame/data/tensor_frame.py", line 78, in __init__
    self.validate()
  File "/home/aki/work/github.com/pyg-team/pytorch-frame/torch_frame/data/tensor_frame.py", line 100, in validate
    raise ValueError(
ValueError: col_names_dict[categorical] must be a list of column names.