Create a Canonical Dataset for Assessing Correctness and Regression Testing of FWI System Calculations

**Describe the task**

Establish tests and data we can use to determine correctness of weather parameters and calculated FWI System values.

**Acceptance Criteria:**
- [ ] Validate interpolated weather inputs using archived 2025 SFMS weather rasters. Need to confirm which stations SFMS used for the interpolation so we can use the same station list in our interpolation.
- [ ] Validate calculated FWI outputs - this can be done with both tabular and raster data (see additional context).
- [ ] Actual scenarios: Reproduce FWI rasters using actual weather observations for the period July 14 to 23, 2025. Starting codes would be from July 13, 2205. Compare against archived SFMS rasters.
- [ ] Confirm how SFMS jobs are triggered and how we can trigger jobs for testing during the winter when SFMS is not operational.

**Additional Context:**
- Tabular test data are provided in Van Wagner and Pickett (1985): https://ostrnrcan-dostrncan.canada.ca/entities/publication/29706108-2891-4e5d-a59a-a77c96bc507c. This test data set is available through the R cffdrs package. The following R script produces FWI System outputs that are within one decimal place of the test data published back in 1985.

library(cffdrs)
data("test_fwi")
output <- fwi(
input = test_fwi,
init = data.frame(ffmc = 85, dmc = 6, dc = 15, lat = 40)
)
output[,c("FFMC","DMC","DC","ISI","BUI","FWI")] <- round(output[,c("FFMC","DMC","DC","ISI","BUI","FWI")], digits = 1)
output[,"DSR"] <- round(output[,"DSR"], digits = 2)
output
- There is also a raster test data set available in the R cffdrs package under the fwiRaster() function.

Start with the provided cffdrs test_fwi example and fwiRaster() reference data, then inspect the archived 2025 SFMS weather and FWI rasters for July 14–23, using July 13 starting codes. Done means the weather interpolation and FWI outputs are compared against the references and the winter job-triggering method is confirmed.

Written by the indexing model from the issue text.