feat: modernize PSO and add convergence research

Migrate the package and examples to the tensor-native PyTorch implementation, add benchmark evidence, and add the guarded post-training convergence protocol with TensorBoard progress monitoring and hash-verified recovery.

Constraint: Preserve one-shot official-test sealing and auditable research artifacts

Rejected: Commit local .omc runs and downloaded datasets | multi-gigabyte runtime state is machine-local

Confidence: high

Scope-risk: broad

Not-tested: Production CUDA run on pieroot-server

pso/_version.py

@@ -0,0 +1,22 @@
+from pathlib import Path
+import sys
+from importlib.metadata import PackageNotFoundError, version
+
+try:
+    __version__ = version("pso2keras")
+except PackageNotFoundError:
+    pyproject_path = Path(__file__).resolve().parent.parent / "pyproject.toml"
+    if pyproject_path.exists():
+        if sys.version_info >= (3, 11):
+            import tomllib
+        else:
+            try:
+                import tomllib  # type: ignore
+            except ImportError:
+                import tomli as tomllib  # type: ignore
+
+        with pyproject_path.open("rb") as f:
+            data = tomllib.load(f)
+            __version__ = data["project"]["version"]
+    else:
+        __version__ = "0.0.0"