Reading delta table using Dataset type File(Machine learning studio)

### Operating System

Windows

### Version Information

N/A

### Steps to reproduce

Folder structure:
/
└── my_data
├── _delta_log
├── part-0000-xxx.parquet
└── part-0001-xxx.parquet
Relative path for dataset_name set up in dataset type(from azure ML v1 APIs) file: my_data/*.parquet

from azureml.core import Workspace,Dataset
import pandas as pd
ws = Workspace.from_config()
dataset = Dataset.get_by_name(ws, "Dataset_name")
downloaded_parquets = dataset.download()
df_list = list()
for file in downloaded_parquets:
df_list.append(pd.read_parquet(file, engine='pyarrow'))
df = pd.concat(df_list)
df.head(6)

### Expected behavior

id name
3 USA
1 FR
2 UK
5 SPAIN

### Actual behavior

id name
1 FR
4 INDIA
5 Crude
2 UK
3 USA
5 SPAIN

The actual data it is showing is all the data that was added, doesn't show the most recent version

### Addition information

_No response_
```[tasklist]
### Tasks
```

Start by reproducing the Python example with the Azure ML Dataset named in the issue and the my_data/*.parquet path. Inspect how Dataset.download handles the folder's Parquet files and _delta_log; done means the resulting dataframe reflects the most recent table version rather than all added rows.

Written by the indexing model from the issue text.