storage blob download-batch fails with "[Errno 20] Not a directory"

### Describe the bug

The `download-batch` command fails when downloaded nested subfolders from Data Lake Storage. Folders, in this case, are represented as zero-byte blobs. For example,

```sh
az storage blob download-batch -s https://sciencelibrarieswestus2.blob.core.windows.net/external-data-share -d stanford --pattern stanford/*
```

When you download nested subfolders, a zero byte file gets written to the destination on local disk, called `dataset_a` in my case. Since that's not a directory, you get the following error when you try to write the files inside that `dataset_a` subfolder.

[Errno 20] Not a directory: 'stanford/stanford/dataset_a/dataset_a_Summary.csv'

You can work around by creating the local directory structure manually and downloading the leave files only, but that's quite tedious.

### Related command

az storage blob download-batch

### Errors

[Errno 20] Not a directory

### Issue script & Debug output

Didn't get a chance to do this.

### Expected behavior

make directories instead of zero-byte files.

### Environment Summary

az --version
azure-cli 2.75.0

core 2.75.0
telemetry 1.1.0

Dependencies:
msal 1.33.0b1
azure-mgmt-resource 23.3.0

Python location '/opt/az/bin/python3'
Config directory '/home/christopherbare/.azure'
Extensions directory '/home/christopherbare/.azure/cliextensions'

Python (Linux) 3.12.10 (main, Jun 24 2025, 10:15:18) [GCC 13.3.0]

### Additional context

_No response_