COMPMID-3887: NEGEMMConvolutionLayer hangs up on num_threads>18

COMPMID-3887: NEGEMMConvolutionLayer hangs up on num_threads>18

When store the Tensor shape information, the size was clamped to 32bit, which cause the memory allocated is not big enough.

Signed-off-by: Sheri Zhang <sheri.zhang@arm.com>
Change-Id: I9f7dfcd5595a143b0ed4f6973e20bcd9d776b673
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/4331
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: Georgios Pinitas <georgios.pinitas@arm.com>