## Problem

S3 downloads were sometimes failing with `NotADirectoryError` and
`FileExistsError` when S3 buckets contained objects with conflicting
naming patterns that cannot be represented in traditional filesystem
hierarchies.

**Example conflict:**
- S3 object: `foo` (file)
- S3 object: `foo/documents` (file requiring foo to be a directory)

This created a race condition where download order determined
success/failure

## Solution

Used tempfile to create unique download paths for each S3 object:

**Before:**
```
S3: "foo" → Local: /downloads/foo
S3: "foo/documents" → Local: /downloads/foo/documents
Conflict: foo cannot be both file and directory
```

**After:**
```
S3: "foo" → Local: /downloads/a1b2c3d4e5f6/foo
S3: "foo/documents" → Local: /downloads/9g8h7i6j5k4l/documents
No conflicts: Each file gets unique directory
```

## Future Work

This PR targets only the s3 downloads. I think it would make sense to
use tempfiles for all downloads (as in [PR
#571](#571)),
but that requires more extensive changes to implement cleanly. This fix
provides immediate relief from the path conflict issues while we work on
the more comprehensive tempfile solution.