Make Shark Optional for the "OTBDimensionalityReductionLearning" module

Summary

Make Shark Optional for the DimensionalityReductionLearning module.

Rationale

The OTBDimensionalityReductionLearning module adds machine learning models for dimensionality reduction :

• The SOMModel : using the Self organizing map classes of the OTB.
• The AutoencoderModel : using the Shark autoencoder classes.
• The PCAModel : using the Shark PCA classes.

Currently, this module can't be built without the Shark library, i.e. we can't build only the SOMModel.

This is actually a problem, because this means that the OTBAppDimensionalityReduction also requires Shark. Therefore if the OTB is built without Shark, the DimensionalityReduction application (that has nothing to do with Shark and the DimensionalityReductionLearning module) will not be built.

This merge request makes Shark Optional for the DimensionalityReductionLearning module, similarly as it is done for the Supervised and Unsupervised modules. If OTB is built without Shark, the TrainDimensionalityReduction will only have the SOM algorithm available.

Also, this MR adds missing copyrights in the Unsupervised module.

Copyright

The copyright owner is CNES and has signed the ORFEO ToolBox Contributor License Agreement.

---

Check before merging:

• All discussions are resolved
• At least 2 votes from core developers, no vote.
• The feature branch is (reasonably) up-to-date with the base branch
• Dashboard is green
• Copyright owner has signed the ORFEO ToolBox Contributor License Agreement