Fix: Verify scales are not None for Cutlass FP8 FusedMoE (#1961)

## 📌 Description
Verify quant scales for fp8 are non null in cutlass FusedMoE path.
Currently, if these tensors are passed as None from python it will
result in segmentation fault.

<!-- This is an auto-generated comment: release notes by coderabbit.ai
-->
## Summary by CodeRabbit

* **Bug Fixes**
* Enhanced validation for FP8 quantization parameters to improve system
robustness and prevent potential null reference issues during
quantization operations, reducing the risk of runtime errors when
processing quantized model data.
<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Signed-off-by: Amir Klein <[email protected]>

Author: amirkl94

csrc/fused_moe/cutlass_backend/flashinfer_cutlass_fused_moe_sm100_binding.cu

@@ -822,6 +822,13 @@ class FusedMoeRunner : public tvm::ffi::ModuleObj {
       auto const fc2_dequant = quant_scales.value()[2];
       auto const fc1_input_dequant = quant_scales.value()[3];
 
+      TVM_FFI_ICHECK(fc1_dequant.get() != nullptr) << "Expecting fc1_dequant to be non null";
+      TVM_FFI_ICHECK(fc2_quant.get() != nullptr) << "Expecting fc2_quant to be non null";
+      TVM_FFI_ICHECK(fc2_dequant.get() != nullptr)
+          << "Expecting fc2_dequant_dequant to be non null";
+      TVM_FFI_ICHECK(fc1_input_dequant.get() != nullptr)
+          << "Expecting fc1_input_dequant to be non null";
+
       // Check types
       CHECK_INPUT_TYPE(fc1_dequant, dl_float32);
       CHECK_INPUT_TYPE(fc2_quant, dl_float32);