[FEA]: Store peak memory bandwidth in tuning database

### Is this a duplicate?

- [x] I confirmed there appear to be no [duplicate issues](https://github.com/NVIDIA/cccl/issues) for this request and that I agree to the [Code of Conduct](CODE_OF_CONDUCT.md)

### Area

Infrastructure

### Is your feature request related to a problem? Please describe.

The tuning database does not store the peak memory bandwith of the GPU where a benchmark was executed. It currently only stores the GPU name, bus width in bits, SM count and whether ECC was on.

Therefore, comparison and plot scripts based on tuning databases cannot report the absolute achieved bandwidth.

### Describe the solution you'd like

Additionally store the peak memory bandwidth.

### Describe alternatives you've considered

_No response_

### Additional context

_No response_