fix: missing header include in decode kernel jit binding (#1802)

## 📌 Description

This commit adds the missing include of the header `decode.cuh` in the
JIT binding `csrc/batch_decode.cu`, without this fix, there will be
linking error when linking `batch_decode.o`, `batch_decode_kernel.o` and
`batch_decode_jit_binding.o`.

The reason is that, `csrc/batch_decode.cu` relies on `scheduler.cuh` for
decoding attention planning. `scheduler.cuh` declares the decode kernel
of `BatchDecodeWithPagedKVCacheKernel`, which is implemented in
`decode.cuh`, but does not include `decode.cuh`. Therefore, if
`csrc/batch_decode.cu` does not include `decode.cuh`, the decode kernel
will be a symbol that is declared but not implemented.


## 🔍 Related Issues

N/A

## 🚀 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

- [x] I have installed `pre-commit` by running `pip install pre-commit`
(or used your preferred method).
- [x] I have installed the hooks with `pre-commit install`.
- [x] 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

N/A

Author: MasterJH5574

csrc/batch_decode.cu

@@ -13,6 +13,7 @@
  * See the License for the specific language governing permissions and
  * limitations under the License.
  */
+#include <flashinfer/attention/decode.cuh>
 #include <flashinfer/attention/scheduler.cuh>
 #include <flashinfer/pos_enc.cuh>
 #include <flashinfer/utils.cuh>