Fix cute dsl gemm API wrong arg name and silent error when passing wrong kwargs (#1619)

<!-- .github/pull_request_template.md -->

## 📌 Description

* fix name error: not "mm" but "mn"
* make it explicitly fail now. o/w e.g. when users need to pass in
`signals` and wrongly use a old flashinfer version, they will not
realize it and get confusing errors.

<!-- What does this PR do? Briefly describe the changes and why they’re
needed. -->

## 🔍 Related Issues

<!-- Link any related issues here -->

## 🚀 Pull Request Checklist

Thank you for contributing to FlashInfer! Before we review your pull
request, please make sure the following items are complete.

### ✅ Pre-commit Checks

- [ ] I have installed `pre-commit` by running `pip install pre-commit`
(or used your preferred method).
- [ ] I have installed the hooks with `pre-commit install`.
- [ ] I have run the hooks manually with `pre-commit run --all-files`
and fixed any reported issues.

> If you are unsure about how to set up `pre-commit`, see [the
pre-commit documentation](https://pre-commit.com/).

## 🧪 Tests

- [ ] Tests have been added or updated as needed.
- [ ] All tests are passing (`unittest`, etc.).

## Reviewer Notes

<!-- Optional: anything you'd like reviewers to focus on, concerns, etc.
-->

---------

Co-authored-by: Avery Yingyi Huang <[email protected]>

Author: fzyzcjy

flashinfer/cute_dsl/blockscaled_gemm.py

@@ -2811,13 +2811,15 @@ def grouped_gemm_nt_masked(
         # Note: only support deepgemm-like shape for now
         k = k * 2
 
-    mma_tiler_mn = kwargs.get("mma_tiler_mm", (128, 128))
-    cluster_shape_mn = kwargs.get("cluster_shape_mm", (1, 1))
+    mma_tiler_mn = kwargs.pop("mma_tiler_mn", (128, 128))
+    cluster_shape_mn = kwargs.pop("cluster_shape_mn", (1, 1))
     if sm_count is None:
         sm_count = get_num_sm(a_torch.device)
 
-    alpha = kwargs.get("alpha")
-    alpha_dtype = kwargs.get("alpha_dtype")
+    alpha = kwargs.pop("alpha", None)
+    alpha_dtype = kwargs.pop("alpha_dtype", None)
+
+    assert len(kwargs) == 0, f"Unsupported kwargs: {kwargs}"
 
     major, minor = get_compute_capability(a_torch.device)
     if major == 11 and minor == 0: