Fix Coverity OVERRUN false positive in _allocate_result

Declare f_ndim as a C int so that f_ndim * sizeof(npy_intp) is computed
as a plain C multiplication instead of being routed through a Python
object. The untyped variable caused Cython to generate a
__Pyx_PyLong_As_size_t conversion whose (size_t)-1 error sentinel led
Coverity to report an out-of-bounds access (OVERRUN, CID 652776/652762)
on the PyMem_Malloc/memcpy of f_shape. This mirrors b_ndim in
_pad_array, which is already typed and was never flagged.

Author: antonwolfy

CHANGELOG.md

@@ -11,6 +11,7 @@ and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0
 ### Changed
 
 ### Fixed
+* Declared `f_ndim` as a C `int` in `_allocate_result` so the buffer size is computed in C rather than through a Python object, resolving a Coverity out-of-bounds (OVERRUN) false positive [gh-364](https://github.com/IntelPython/mkl_fft/pull/364)
 
 ## [2.3.2] - 2026-08-04
 

mkl_fft/_pydfti.pyx

@@ -302,6 +302,7 @@ cdef cnp.ndarray _allocate_result(
     cdef cnp.npy_intp *f_shape
     cdef cnp.ndarray f_arr "ff_arrayObject"
     cdef int x_arr_is_fortran
+    cdef int f_ndim
 
     f_ndim = cnp.PyArray_NDIM(x_arr)