Add more energy calibration methods

[RLumSK]

This is a more complicated PR. It boils down to the idea that each dose-rate calibration is based on spectra that have (or don't have) one or more energy/channel calibrations. Users try all kinds of weird things, and I realised that we cannot take for granted that all spectra are energy-calibrated. On the other hand, since the energy calibration likely did not change, it would be nice if dose_predict() were to take over the calibration from the original dataset.

Description

• ad code that forwards the energy calibration model during the call of dose_fit() and store it in the info slot
• ad consideration of this energy calibration to dose_predict()
• up show method for the calibration curve
• ad a lot more tests
• ad NEWS

Side effect of this change is that, because it was necessary to touch the class methods, that energy_calibrate() can now copy over an existing calibration from another spectrum.

Related Issue

Not applicable.

Example

data("clermont")
spc_dir <- system.file("extdata/BDX_LaBr_1/calibration", package = "gamma")
spc <- read(spc_dir)
bkg_dir <- system.file("extdata/BDX_LaBr_1/background", package = "gamma")
bkg <- read(bkg_dir)
doses <- clermont[, c("gamma_dose", "gamma_error")]

dose_fit(spc, bkg, doses,  range_Ni = c(300, 2800), range_NiEi = c(165, 2800))

Calibration curve:
* authors: 
* date: 2024-09-20 11:05:06.380417
* energy_calibration: NA
Warning message:
All spectra without energy calibration. You can proceed but it is not recommended! 

dose_predict(calib, spc_no_calib)

No energy calibration found for 'spectrum', apply calibration from dose-rate calibration model!

[RLumSK]

@nfrerebeau I had updated the snapshots but GA tell otherwise. I just merge and will check again after.