```python
import torch
import torch.nn as nn
from torch.utils.cpp_extension import load_inline

_cpp_source = """
#include <torch/extension.h>
torch::Tensor fused_activations_cuda(torch::Tensor x, torch::Tensor add);
"""

_cuda_source = r"""
#include <torch/extension.h>
#include <ATen/OpMathType.h>
#include <ATen/cuda/CUDAContext.h>
#include <c10/cuda/CUDAGuard.h>
#include <c10/cuda/CUDAException.h>
#include <cuda_runtime.h>

template <typename T>
__device__ __forceinline__ T activate(T x) {
    x = x / (T(1) + exp(-x));
    x = tanh(x);
    x = T(0.5) * x * (T(1) + erf(x * T(0.7071067811865475244)));
    return x < T(-1) ? T(-1) : (x > T(1) ? T(1) : x);
}

template <>
__device__ __forceinline__ float activate<float>(float x) {
    x = x / (1.0f + __expf(-x));
    x = tanhf(x);
    x = 0.5f * x * (1.0f + erff(x * 0.7071067811865475244f));
    return fminf(1.0f, fmaxf(-1.0f, x));
}

template <typename scalar_t>
__global__ void fused_activations_kernel(
    const scalar_t* __restrict__ x,
    const scalar_t* __restrict__ add,
    scalar_t* __restrict__ out,
    int64_t n,
    int64_t columns
) {
    using compute_t = at::opmath_type<scalar_t>;
    const int64_t base =
        static_cast<int64_t>(blockIdx.x) * blockDim.x * 4 + threadIdx.x;

    #pragma unroll
    for (int k = 0; k < 4; ++k) {
        const int64_t i = base + k * blockDim.x;
        if (i < n) {
            const compute_t value =
                static_cast<compute_t>(x[i]) +
                static_cast<compute_t>(add[i % columns]);
            out[i] = static_cast<scalar_t>(activate<compute_t>(value));
        }
    }
}

torch::Tensor fused_activations_cuda(torch::Tensor x, torch::Tensor add) {
    TORCH_CHECK(x.is_cuda() && add.is_cuda(), "Expected CUDA tensors");
    TORCH_CHECK(x.device() == add.device(), "Device mismatch");
    TORCH_CHECK(x.scalar_type() == add.scalar_type(), "Dtype mismatch");
    TORCH_CHECK(x.is_contiguous() && add.is_contiguous(),
                "Expected contiguous tensors");
    TORCH_CHECK(x.dim() >= 1 && add.dim() == 1 &&
                x.size(-1) == add.numel(), "Invalid add shape");

    const c10::cuda::CUDAGuard guard(x.device());
    auto out = torch::empty_like(x);
    const int64_t n = x.numel();
    if (n == 0) return out;

    constexpr int threads = 256;
    const int blocks = static_cast<int>((n + threads * 4 - 1) / (threads * 4));
    const auto stream = at::cuda::getCurrentCUDAStream();

    AT_DISPATCH_FLOATING_TYPES(x.scalar_type(), "fused_activations_cuda", [&] {
        fused_activations_kernel<scalar_t><<<blocks, threads, 0, stream>>>(
            x.data_ptr<scalar_t>(),
            add.data_ptr<scalar_t>(),
            out.data_ptr<scalar_t>(),
            n,
            x.size(-1)
        );
    });
    C10_CUDA_KERNEL_LAUNCH_CHECK();
    return out;
}
"""

_fused_ops = load_inline(
    name="linear_swish_tanh_gelu_hardtanh_fused_v1",
    cpp_sources=_cpp_source,
    cuda_sources=_cuda_source,
    functions=["fused_activations_cuda"],
    extra_cflags=["-O3"],
    extra_cuda_cflags=["-O3"],
    verbose=False,
)


class ModelNew(nn.Module):
    def __init__(self, in_features, out_features, add_value_shape):
        super().__init__()
        self.matmul = nn.Linear(in_features, out_features)
        self.add_value = nn.Parameter(torch.randn(add_value_shape))

    def forward(self, x):
        x = self.matmul(x)
        if (
            x.is_cuda
            and x.dtype in (torch.float32, torch.float64)
            and self.add_value.device == x.device
            and self.add_value.dtype == x.dtype
            and self.add_value.ndim == 1
            and self.add_value.numel() == x.shape[-1]
        ):
            return _fused_ops.fused_activations_cuda(
                x.contiguous(), self.add_value.contiguous()
            )

        x = x + self.add_value
        x = torch.sigmoid(x) * x
        x = torch.tanh(x)
        x = torch.nn.functional.gelu(x)
        return torch.nn.functional.hardtanh(x, min_val=-1, max_val=1)
```