Add test_DCSAM.py and HybridInference.ipynb example notebook

[Copilot]

Adds Python-level coverage for DCSAM and a user-facing hybrid inference tutorial, both missing from the current codebase.

python/gtsam/tests/test_DCSAM.py

Five unit tests translated from gtsam/hybrid/tests/testDCSAM.cpp:

Test	What it covers
test_constructor	Default and ISAM2Params-based construction
test_simple_discrete	Pure discrete MPE via DecisionTreeFactor
test_simple_mixture_factor	HybridNonlinearFactor with log-normalisation constants; verifies tighter-prior mode wins
test_simple_slam_batch	8-pose Pose2 SLAM batch solve; checked against C++ reference values
test_simple_slam_incremental	Same trajectory added one pose at a time

dcsam = DCSAM()
initial_guess = HybridValues(VectorValues(), DiscreteValues(), initial_continuous)
dcsam.update(graph, initial_guess)
dcvals = dcsam.calculateEstimate()
self.gtsamAssertEquals(dcvals.nonlinear(), expected, 1e-5)

python/gtsam/examples/HybridInference.ipynb

Three-part tutorial notebook:

1. Discrete-only — DecisionTreeFactor MPE via DCSAM
2. Continuous-only SLAM — DCSAM as thin iSAM2 wrapper for Pose2 odometry
3. True hybrid inference — binary motion mode (normal vs. slippery surface) via HybridNonlinearFactor; demonstrates joint pose + discrete mode estimation with matplotlib visualisation

---

📱 Kick off Copilot coding agent tasks wherever you are with GitHub Mobile, available on iOS and Android.

[dellaert]

@copilot the discrete only and continuous only examples are of less interest. I like the octagon, though. Can we just create one example, an octagon with switching motion models?

[Copilot]

@dellaert The model claude-sonnet-4.6 is not available for your account. This can happen if the model was disabled by your organization's policy or if your Copilot plan doesn't include access to it.

You can try again without specifying a model (just @copilot) to use the default, or choose a different model from the model picker.

If you want to contact GitHub about this error, please mention the following identifier so they can better serve you: 04de8f4c-33af-4c02-a326-fde0cdb59780

[dellaert]

the discrete only and continuous only examples are of less interest. I like the octagon, though. Can we just create one example, an octagon with switching motion models?