Removed **kwds from parse function in read_excel

[pacificdragon]

• closes BUG: xl.parse index_col ignoring skiprows #50953 (Replace xxxx with the GitHub issue number)
• Tests added and passed if fixing a bug or adding a new feature
• All code checks passed.
• Added type annotations to new arguments/methods/functions.
• Added an entry in the latest doc/source/whatsnew/vX.X.X.rst file if fixing a bug or adding a new feature.

The code works fine locally.

Checking CI/CD pipeline now.

[phofl]

This would need tests, but apart from that this is an API change, not sure if we can do this without a deprecation cycle

[rhshadrach]

In the comment below, I'm assuming that all valid kwargs that can be passed to parse are still available, but haven't confirmed that yet.

this is an API change, not sure if we can do this without a deprecation cycle

Part of this change is raising if a user passes invalid keyword args. I think we can regard this as a bugfix. I guess this change might also mess up someone who is using positional arguments - I'd have to guess that is quite rare but I suppose that aspect is an API change. Would you agree?

[phofl]

No problem with making the keywords explicit, but would prefer keeping the order, also would rather deprecate the kwargs instead of removing outright, like we did in other places with unused kwargs

[rhshadrach]

I think the way forward is to emit a FutureWarning if invalid kwargs are in present. I assume @pacificdragon added the keyword arguments in their current locations to agree with the ordering of other methods, but I haven't checked this. We currently can't add new arguments unless they are at the end to retain backwards compatibility. For this, I think there are two ways forward:

a. Add keywords to the tail end of the signature
b. Deprecate specifying any argument other than sheet_name by position

If we go with (a), we'll still have different ordering of arguments which isn't great. With (b), we can add arguments in any order after the deprecation is enforced. So I'd lean toward (b) here unless it's preferred that these arguments are allowed to be positional.

[phofl]

b sounds good, also +1 on adding a FutureWarning and removing in 3.0

[WillAyd]

I would actually be OK with doing this without a deprecation cycle. I think it is a bug for a function to silently accept arguments it does nothing with. We also have a precedent for doing this in the excel space like #34464

[pacificdragon]

Thanks, Everyone for taking action on this. @rhshadrach Created PR for the same 51331