DeepSpeed

7af561c2 - Fix illegal memory access with multi_tensor_apply size above INT_MAX (#7639)

Fix illegal memory access with multi_tensor_apply size above INT_MAX (#7639) This change fixes an overflow issue in `TensorListMetadata` where the `sizes` array used `int` (32-bit signed integer). This caused incorrect behavior (e.g., no parameter updates) when handling tensor sizes exceeding `INT_MAX` (2^31 - 1). The change here is identical to NVIDIA/apex PR [#1825](https://github.com/NVIDIA/apex/pull/1825) for `multi_tensor_apply.cuh`. For further details regarding this fix, please refer to issue [#7640](https://github.com/deepspeedai/DeepSpeed/issues/7640). Signed-off-by: Wang Yan <wangyan.blaze@bytedance.com>