Track global_amax for weight FP4 MSE sweep; Refactor to NVFP4StaticQantizer, NVFP4MSECalibrator (#849)

**Type of change:** ? <!-- Use one of the following: Bug fix, new
feature, new example, new tests, documentation. -->

**Overview:** ?

<!-- You can potentially add a usage example below. -->

```python
```

<!-- Mention how have you tested your change if applicable. -->

<!-- If you haven't finished some of the above items you can still open
`Draft` PR. -->

- **Make sure you read and follow [Contributor
guidelines](https://github.com/NVIDIA/Model-Optimizer/blob/main/CONTRIBUTING.md)**
and your commits are signed.
- **Is this change backward compatible?**: Yes/No <!--- If No, explain
why. -->
- **Did you write any new necessary tests?**: Yes/No
- **Did you add or update any necessary documentation?**: Yes/No
- **Did you update
[Changelog](https://github.com/NVIDIA/Model-Optimizer/blob/main/CHANGELOG.rst)?**:
Yes/No <!--- Only for new features, API changes, critical bug fixes or
bw breaking changes. -->

<!-- E.g. related issue. -->

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

* **New Features**
* Added NVFP4StaticQuantizer for improved 4-bit quantization with
enhanced precision control
* Introduced NVFP4MSECalibrator with flexible candidate generation for
calibration optimization

* **Improvements**
* Optimized GPU kernels for Hopper+ graphics cards with better
performance
  * Extended Triton support to broader GPU compatibility
* Enhanced backward compatibility for restoring previously quantized
models

* **Tests**
* Added comprehensive test coverage for new quantizers and calibration
methods

<!-- end of auto-generated comment: release notes by coderabbit.ai -->

---------

Signed-off-by: realAsma <akuriparambi@nvidia.com>
Signed-off-by: Suguna Velury <178320438+sugunav14@users.noreply.github.com>

Author: realAsma

modelopt/torch/quantization/triton/__init__.py

@@ -34,6 +34,10 @@
         from .fp4_kernel import *
         from .fp8_kernel import *
 
+        # fp4_kernel_hopper requires compute >= 8.9 (uses tl.float8e4nv)
+        if torch.cuda.get_device_capability() >= (8, 9):
+            from .fp4_kernel_hopper import *
+
         # fp4_kernel_hopper requires compute >= 8.9 (uses tl.float8e4nv)
         if torch.cuda.get_device_capability() >= (8, 9):
             from .fp4_kernel_hopper import *