Add test data for quantile mapping acceptance tests

[maxwhitemet]

This PR adds the test data associated with the acceptance tests for #2264.

[gavinevans]

Thanks @maxwhitemet 👍

I've added a couple of comments.

[gavinevans]

The units for all these files seem to be mm/hr, but the values in the files look like m/s.

[gavinevans]

These acceptance test inputs are overly small and look like they're just the same as the unit test data. The aim with the acceptance test inputs is to provide something a bit more realistic. A larger domain (50x50 or 100x100) would be helpful to understand the behaviour. It would be helpful to add visualisations of all the inputs and kgos to the ticket.

The current data looks like as below. In this example, it looks like the forecast to larger than the reference so quantile mapping is scaling down the values.

[maxwhitemet]

The below images demonstrate that quantile mapping has been appropriately applied.

• The forecast bias towards lower precipitation rates has been corrected, with the CDF of the transformed forecast aligning well with the reference data
• Values below 2 mm h-1 (intentionally high for illustration purposes) from the biased forecast have been maintained in the transformed forecast
• The spatial characteristics of the biased forecast have been maintained in the transformed forecast