Tensorboard logging breaks with certain scalar values with numpy >= 2.4.0

Nobody has claimed this yet.

### Bug description

As of [Numpy 2.4.0](https://numpy.org/doc/stable/release/2.4.0-notes.html#raise-typeerror-on-attempt-to-convert-array-with-ndim-0-to-scalar), converting an 0-dimensional array to a scalar is now a TypeError, following the expiration of a lengthy deprecation period.

This is triggered in [src/lightning/fabric/loggers/tensorboard.py](https://github.com/Lightning-AI/pytorch-lightning/blob/0a0f0610a4d223a258cd73e65abe852a8f703226/src/lightning/fabric/loggers/tensorboard.py#L198-L218) in several tests when we run with numpy >= 2.4. It is not clear to me why the CI is not picking this up, though, since the `requirements/fabric/test.txt` file does seem to allow more recent versions of numpy.

### What version are you seeing the problem on?

master

### Reproduced in studio

_No response_

### How to reproduce the bug

```python

```

### Error messages and logs

```
# Error messages and logs here please
```

### Environment

Current environment

```
#- PyTorch Lightning Version (e.g., 2.5.0):
#- PyTorch Version (e.g., 2.5):
#- Python version (e.g., 3.12):
#- OS (e.g., Linux):
#- CUDA/cuDNN version:
#- GPU models and configuration:
#- How you installed Lightning(`conda`, `pip`, source):
```

### More info

_No response_

cc @ethanwharris @lantiga

Start in src/lightning/fabric/loggers/tensorboard.py around lines 198-218 and inspect the scalar logging path that fails with NumPy 2.4. Check the related tests and requirements/fabric/test.txt, then run the affected tests with NumPy 2.4 or newer. Done means scalar logging works for the affected values and CI exercises the supported NumPy version.

Written by the indexing model from the issue text.