Fixed #10785: Modified berkowitz_eigenvals() function

[aravindkanna]

This fixes the issue 10785.

[aravindkanna]

As the empty matrix have no minors, I think it would be better to return an empty tuple for the berkowitz_minors() function.

[aravindkanna]

In [87]: a.berkowitz_eigenvals()
Out[87]: {}
In [88]: a.berkowitz_minors()
Out[88]: ()

Now this would be the output for two modified functions(a is an Empty Matrix).

[aravindkanna]

@moorepants , sir please review this

[aravindkanna]

No output has been received in the last 10 minutes, this potentially indicates a stalled build or something wrong with the build itself.
what is this?

[leosartaj]

240 commits seem to be a lot. Can you update this with master? So only latest commits are shown.

[aravindkanna]

@leosartaj , thank you 😄 can this be in?

[jksuom]

Is this test necessary after berkowitz has been modified to work with an empty matrix?

[aravindkanna]

yeah, sorry. It is not necessary. I have created this before that. Thank you @jksuom . Can I close this now?

[jksuom]

It is not necessary to totally close the PR, the tests could remain.
This is what I get with the current master for M = Matrix():

In [3]: M.berkowitz_minors()
Out[3]: (1,)
In [4]: M.berkowitz_eigenvals()
Out[4]: {}

I think you could leave these tests in this PR (with the first one properly edited). On the other hand, there seems to be no reason for special treatment of empty matrices in berkowitz_minors and berkowitz_eigenvals.
So the contents of this PR could consist of those tests alone.

[jksuom]

Looks good, thanks.