fix: fsspec closure checks

[ariostas]

This PR fixes #1292.

I fixed the closed property of FSSpecSource so that it doesn't always return False. I asked Luis about this, and here's my current understanding. This property was added so that it didn't just use the inherited property from Source since it wouldn't work for custom executors.

uproot5/src/uproot/source/chunk.py

Lines 168 to 177 in a331467

	@property
	def closed(self) -> bool:
	"""
	True if the associated file/connection/thread pool is closed; False
	otherwise.
	"""
	if self._executor is None:
	return True
	return self._executor.closed

However, we can simply use the closed property of _file as shown in the changes below. I'm tagging @lobis to see if he has any concerns about this.

I also changed the chunk(s) methods to perform a closure check before reading. So now the example in #1292 fails.

>>> import uproot
>>> import skhep_testdata
>>> with uproot.open(skhep_testdata.data_path("uproot-Zmumu.root")) as file:
...     tree = file["events"]
...     
>>> tree["px1"].array()
Traceback (most recent call last):
  File "<python-input-3>", line 1, in <module>
    tree["px1"].array()
    ~~~~~~~~~~~~~~~~~^^
  File "/Users/ariostas/Documents/uprootWork/uproot5/src/uproot/behaviors/TBranch.py", line 1825, in array
    _ranges_or_baskets_to_arrays(
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^
        self,
        ^^^^^
    ...<9 lines>...
        interp_options,
        ^^^^^^^^^^^^^^^
    )
    ^
  File "/Users/ariostas/Documents/uprootWork/uproot5/src/uproot/behaviors/TBranch.py", line 3026, in _ranges_or_baskets_to_arrays
    hasbranches._file.source.chunks(ranges, notifications=notifications)
    ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/Users/ariostas/Documents/uprootWork/uproot5/src/uproot/source/fsspec.py", line 136, in chunks
    raise OSError(f"memmap is closed for file {self._file_path}")
OSError: memmap is closed for file /Users/ariostas/.local/skhepdata/uproot-Zmumu.root

[jpivarski]

Thanks! This is the right thing to do, but I think the text of the error message (below) was wrong. Unless you happen to know that it's really memmap, please change that.

[ariostas]

Thank you, @jpivarski! I fixed the error messages.

[jpivarski]

This is a simple enough PR that I don't think you have any more changes coming, so I'll merge now. Thanks!