in CSV most common separator is the comma but sometimes other data contains a comma which I know the workaround is quotes but if it was a random thing like: ;%#£ then the likelihood of this appearing in the data is substantially smaller than a "," appearing.
We will see if there is interest to add this feature. In the meantime, if you have combination of characters that you would like to use as a separator, you could use Search/replace block to replace ;%#£ with a character that doesn't appear in your data, then use it as a separator.