Inconsistent file transformations from combine file in Dataflow

I'm not sure how to get the Power Query code for each step. But I've copied the formulas for the steps:

(Here, I'm using the base URL for the sharepoint folder as the source, then filtering for the specific subfolder and for the latest file created)

SharePoint.Files("https://contoso.sharepoint.com/sites/folder", [ApiVersion = 15])Table.SelectRows(Source, each [Folder Path] = "https://contoso.sharepoint.com/sites/folder/subfolder/CSV Files/")
Table.Sort(#"Filtered rows", {{"Date created", Order.Descending}})Table.FirstN(#"Sorted rows", 1)Table.SelectRows(#"Kept top rows", each [Attributes]?[Hidden]? <> true)

The below steps are automatically created by the dataflow when I "combine file" - e.g. expand the 1 latest created file resulting from the filtering done above
Table.AddColumn(#"Filtered hidden files", "Transform file", each #"Transform file"([Content]))
Table.RenameColumns(#"Invoke custom function", {{"Name", "Source.Name"}})
Table.SelectColumns(#"Renamed columns", {"Source.Name", "Transform file"})

(this is where dataflow expands the file into a table)
Table.ExpandTableColumn(#"Removed other columns", "Transform file", Table.ColumnNames(#"Transform file"(#"Sample file")))

Table.TransformColumnTypes(#"Choose columns", {{"A", type text}, {"B", type text}, {"C", type text}, {"D", type text}, {"E", type datetime}, ... , {"T", type text}})

In the sample code above and when I first created this dataflow, there were only 20 columns in the original CSV file (A -> T). I refreshed the dataflow with the new CSV file for this month - which has 21 columns (for example: A -> ... -> J, J1 -> ... -> T), and the resulting columns from the "ExpandTableColumn" are A -> ... -> J, J1 -> ... -> S (no T).

======================================================

I wanted to add that I looked into the "Queries" panel of the dataflow more carefully, at "Transform file from Query" > "Helper Queries" > (function) "Transform File". The function in the "Transform file" is:

(Parameter as binary) => let

Source = Csv.Document(Parameter, [Delimiter = ",", Columns = 20, QuoteStyle = QuoteStyle.None]),

#Promoted headers" = Table.PromoteHeaders(Source, [PromoteAllScalars = true])

in

#"Promoted headers"

So I think I found the "internal" logic that limits the column count to 20. Is there a way I can set this number to the variable total number of columns detected in the file - rather than a set number?