[source-sftp-bulk] Data type for columns always inferred as 'string' when parsing CSVs

### Connector Name

source-sftp-bulk-zip

### Connector Version

1.81.1

### What step the error happened?

During the sync

### Relevant information

The connectorSpecification for this connector contains the hidden field "inference_type" that defaults to None:
```
# rest of the connector specification omitted
inference_type": {
"title": "Inference Type",
"description": "How to infer the types of the columns. If none, inference default to strings.",
"default": "None",
"airbyte_hidden": true,
"enum": [
"None",
"Primitive Types Only"
]
},
... # snip
```
As a result, data type for all columns [is always inferred as 'string'](https://github.com/airbytehq/airbyte-python-cdk/blob/d3c94190e12864b5ecb918d9ad3cc5bd29514d16/airbyte_cdk/sources/file_based/file_types/csv_parser.py#L188-L194) when parsing CSV files .

Is this behaviour intentional?

### Contribute

- [x] Yes, I want to contribute

---
**Internal Tracking:** https://github.com/airbytehq/oncall/issues/11138