[OptApp][SIApp] Clamper projection

[talhah-ansari]

📝 Description

This PR migrates the SmoothClamper utility from the SystemIdentificationApplication to the OptimizationApplication, refactors the Python bindings and usage accordingly, and introduces a ClampingProjection for controlling and projecting variable bounds. It also updates the relevant tests and integrates the new projection utility into the control workflow.

🆕 Changelog

• Moved smooth_clamper.cpp, smooth_clamper.h, and their test from SystemIdentificationApplication to OptimizationApplication.
• Updated Python bindings to expose SmoothClamper only in OptimizationApplication, removing it from SystemIdentificationApplication and adding it to the new location.
• Added a new ClampingProjection class in opt_projection.py to generalize clamping/projection of design variables, with helper class TensorAdaptorBoundingManager for interval management. Registered clamping_projection as a supported projection type.
• Moved class TensorAdaptorBoundingManager from tensor_adaptor_utils.py in SystemIdentificationApplication to a helper class in opt_projection.py because it is the only place it's being used and thus avoids dependency of OptApp on SIApp.
• Refactored data_values_control.py to use the new projection utility instead of directly managing SmoothClamper and interval bounding.
• Updated and moved the test_smooth_clamper test to the new location and ensured it is included in the test suite for OptimizationApplication.
• Adjusted the initialization order in optimization_analysis.py to ensure processes are created and initialized in Initialize().

[talhah-ansari]

@sunethwarna I moved TensorAdaptorBoundingManager from tensor_adaptor_utils.py in SIApp to opt_projection.py in OptApp as a helper class, since it's used only there, so we avoid SIApp's dependency on OptApp. Let me know if we need to revert it.