Cascading exceptions not properly reported

Consider the following sample code:

try:
    assert False
except Exception as ex:
    0/0

CPython reports both exceptions:

Traceback (most recent call last):
  File ".\ex_cascade.py", line 2, in <module>
    assert False
AssertionError

During handling of the above exception, another exception occurred:

Traceback (most recent call last):
  File ".\ex_cascade.py", line 5, in <module>
    0/0
ZeroDivisionError: division by zero

IronPython, however, fails to report the original exception:

Traceback (most recent call last):
  File ".\ex_cascade.py", line 5, in <module>
ZeroDivisionError: Attempted to divide by zero.

Note that IronPython has already code to properly display chained exceptions, however the exceptions from the example above are not being chained. Perhaps this is related to IronLanguages/ironpython2#493.

Start with the exception-reporting code in Src/IronPython/Runtime/PythonContext.cs around the linked chained-exception logic, then trace how exceptions raised while handling another exception are propagated. Compare the sample's IronPython and CPython output, and review IronLanguages/ironpython2#493 for related context. Done means the original and subsequent exceptions are both reported with the expected chaining wording.