CSV output formatting fix

[rbiseck3]

If output type is text/csv, wrap in PlainTextResponse to maintain formatting

[rbiseck3]

Not sure what the best response formatting would be when there are multiple files that need csv outputs. Right now it still defaults to the application/json response type which wraps the strings into a json array.

[rbiseck3]

For multi file outputs, the csv content is joined with the others using an outer join so if there were different columns, they would all be persisted.

[rbiseck3]

Since the code in the prepline_test_project directory is autogenerated, I felt it didn't make sense to require a code coverage percentage here since it makes sense that there might be more code generated than needed for every case of wrapped code from a notebook. That directory was removed from the make target that runs pytest.

[awalker4]

I'm getting an error with make tidy-notebooks, which is going to execute all the cells. Is there a missing pandas import somewhere?

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
Cell In[2], line 1
----> 1 print(pipeline_api("some text", "text/csv", "isd"))

Cell In[1], line 7, in pipeline_api(text, response_type, response_schema)
      2 def pipeline_api(
      3     text,
      4     response_type="text/csv",
      5     response_schema="isd",
      6 ):
----> 7     data = pd.DataFrame(data={"silly_result": [str(len(text)), text, str(response_type), str(response_schema)]})
      8     if response_type == "text/csv":
      9         return data.to_csv()

NameError: name 'pd' is not defined
NameError: name 'pd' is not defined