Function roi_pool 

pub fn roi_pool(
    features: &Tensor,
    rois: &[(f32, f32, f32, f32)],
    output_size: (usize, usize),
) -> Result<Tensor, DetectError>

RoI Pooling: for each RoI, divides the region into an output_size grid and max-pools each cell.

• features – [H, W, C] single-image feature map.
• rois – slice of (x1, y1, x2, y2) in feature-map coordinates.
• output_size – (out_h, out_w).

Returns a tensor of shape [num_rois, out_h, out_w, C].