Clear pending Python error when a Try-style conversion fails

[jhonabreul]

What does this implement/fix? Explain your changes.

PyObject.TryAsManagedObject (and therefore TryAs<T>/TrySafeAs<T>) converted with setError: true and returned false on failure, leaving the Python error indicator set on the thread with no one left to surface it. Callers of a Try* API only observe the boolean, so the stale error survived on the thread state and was thrown by the next unrelated Python call on that thread that checks the error indicator.

This was previously masked because the outermost Py.GIL scope deleted the thread state (and its pending error) on dispose. Since #137 pins thread states for the lifetime of the run, the leaked error persists and poisons subsequent calls on the same thread. This is the root cause of the PythonUtilTests.BothInheritedAndNonInheritedClassesWork(True) failure in QuantConnect/Lean#9614's CI: ExtensionsTests.PyObjectTryConvertToNonDelegateFail exercises a failed TryAs<int> on a Python function, leaving TypeError: int() argument must be a string, a bytes-like object or a real number, not 'function' pending on the NUnit worker thread, which PyModule.FromString in the next Python-touching test then throws. Reproduced deterministically by running both fixtures on a single NUnit worker; fixed with this change.

The fix: TryAsManagedObject now converts with setError: false and clears anything a conversion sub-path may have left pending before returning false. AsManagedObject keeps converting with setError: true so the fetched error still becomes the InvalidCastException cause (fetching consumes the indicator, so that path never leaked).

Also bumps the version to 2.0.63 (package <Version>, AssemblyVersion/AssemblyFileVersion, perf-test baseline reference).

A follow-up commit extends the same guarantee to the converter's collection and label paths, which could leak a pending error even with setError: false: ToList now handles a failed PyObject_GetIter like ToArray does (also fixing a leaked iterator reference); MakeList now fails the conversion when the iterator raises mid-iteration instead of returning a silently truncated list, and clears swallowed errors when not setting errors; and the type_error/overflow labels clear pending errors when not setting errors, mirroring convert_error. The unconditional raises in the implicit-conversion catches are kept on purpose — MethodBinder.Invoke surfaces a pending error as the binding failure reason (covered by ImplicitConversionErrorHandling).

Does this close any currently open issues?

Fixes the PythonUtilTests.BothInheritedAndNonInheritedClassesWork CI failure in QuantConnect/Lean#9614.

Any other comments?

Regression tests added:

• RaisingIteratorFailsArrayConversion — a raising iterator fails the conversion in both setError modes, leaving an error pending only when setError is true.
• OverflowConversionOnlyLeavesErrorWhenSettingErrors — overflowing numeric conversions leave an error pending only when setError is true.
• FailedTryAsDoesNotLeavePythonErrorSet — a failed TryAs<int>/TryAsManagedObject on a Python function leaves no pending error and subsequent PyModule.FromString calls work.
• FailedAsManagedObjectRaisesWithConversionErrorAsCause — AsManagedObject still raises InvalidCastException with the Python conversion error as its cause and no pending error afterwards.

Verified end-to-end against Lean: with 2.0.62, running Common.Util.ExtensionsTests + Common.Util.PythonUtilTests on a single NUnit worker reproduces the exact CI failure; with this build (2.0.63) all 352 tests pass. The full pythonnet embedding test suite passes (960 passed, 8 skipped). A full Lean test-suite sweep with a per-test PyErr-leak detector found this TryAs path to be the only leak source across 36,316 tests.

Checklist

Check all those that are applicable and complete.

• Make sure to include one or more tests for your change
• If an enhancement PR, please create docs and at best an example
• Ensure you have signed the .NET Foundation CLA
• Add yourself to AUTHORS
• Updated the CHANGELOG