Bug fix NDArrayPool and New make parameter db file

[guirodrigueslima]

In this request, a correction was made to the SpecsAnalyser::specsAnalyserTask function, when inserting values ​​into the pNDImage pointer in Snapshot mode, a segmentation fault occurred, the solution was to comment the pNDImage in for() and insert the array once.

During code studies, I noticed that the SpecsAnalyser::setupEPICSParameters function was creating dedicated parameters for each Phoibos device, but the .db file itself did not exist, so a python script was introduced to generate the respective .db file. This script was based on the GenICan script.

Thanks.

[guirodrigueslima]

The commit "Supports logical variable spectrum" was recently included. The latest versions of the Prodigy software have a new feature called Logical Variable Spectrum (LVS). This mode takes multiple snapshots by scanning a type of manipulator (Remote Control Protocol manual, section 2.7, version 1.19, 2022).

During commissioning, we tried to use Snapshot mode and scan the manipulator (ShiftX) manually via script. We noticed that the time doubled compared to Logical Variable mode. For this reason, we decided to integrate LVS mode into the IOC. In this mode, the data received is three-dimensional (section 2.19), so it was necessary to adapt the code without compromising the previously implemented modes (snapshot, fixed transmission, fixed retarding ratio, and fixed energy).

We did a new commissioning using three modes, logical variable, snapshot, and fixed transmission, respectively. Everything worked very well. We also did experiments with exposures (>1), in which the IOC integrates the data automatically.

I will attach some data (.hdf5) from ARPES for different modes (LVS, SFAT and FAT) with Cu target, in the image you can see the material surface states.

test.zip

Thanks.

[guirodrigueslima]

I've split this pull request into: #8 and #9.