added binary_cross_entropy loss and test to paddlepaddle frontend

[diksha-shrivastava13]

Close #15950
A work in progress.

[diksha-shrivastava13]

Hi @Ookamice! Can you kindly review this pull request?
I was a bit doubtful about reshaping, did that since paddlepaddle has this 0-dim and 1-dim differentiation issue.
I had to reorganize the tests a bit as well and move everything from test_nn/test_functional to test_nn and remove test_functional in order to have get_mod_submod_test from run_tests_pr.py to not break down.

Please let me know if I need to improve anything.
Thanks!

[Ookamice]

@diksha-shrivastava13
Hiya, thank you for your contributions. I understand that our testing structure with paddlepaddle isn't up to standard and is great that you tried to resolve it through some minor refactoring. However, could I ask you to open up an issue regarding this and revert those file movements for this PR? This is just for organisation and you can feel free to tackle the new issue you have opened! 👍 (Ideally a PR should only have 2 files changes, that being the new frontend function and the respective test)

Once this is done, I will have a closer look at the logic by manually running tests (as the intelligent-tests-pr isn't picking up on the ones you have written)

Let me know if you have any other questions and concerns.

[diksha-shrivastava13]

Hey @Ookamice! I reverted the movement commit as you asked.
I've let the test_paddle_loss movement as it is since intelligent-tests-pr was throwing a key error of "test_functional/test_functional" in run_tests_pr earlier with the previous location, so it's currently in test_paddle/test_nn instead of test_paddle/test_nn/test_functional where it was previously.
I will open an issue to restructure paddle.nn.functional tests. Thanks for guiding!

[diksha-shrivastava13]

I tried refactoring to see if intelligent-tests-pr picks up tests, but no change.

[Ookamice]

@diksha-shrivastava13
Hiya, thank you for the quick changes. I'd suggest you to ignore the intelligent-pr tests for now as they are still a wip for paddlepaddle functions I believe. I'll run the tests manually through command line in a github codespace to test your code so please could you revert those file movements for this PR. This just just for organization as ideally a single PR solves a single problem 😃

[diksha-shrivastava13]

my pycharm is malfunctioning, none of the changes are getting reverted. i'm going to close this pull request for now and work with a new branch. thanks for helping!