Bias segfault workaround

[micmelesse]

• this pr prevents an issue with a large bias tensor leading to segfaults on some configs due to pointer arithmetic overflow. Triton is unable to address tensors with elements larger than int32 max
• we add an assert to the op to catch cases in the future.
• the solution is to split the forward test in to one with bias and one with out with appropritate config

[vgokhale]

What code does this part of the description:

this pr works around an issue with bias leading to segfaults on some configs.

Might be good to add an assert if the size of the bias tensor is > 2^31.

[micmelesse]

@vgokhale Bad wording on my part. I have updated the description and the assert.

[groenenboomj]

The core dump mentioned in the removed skip was from the issue fixed here right? Do we still need to run a separate UT over parameterizing bias?

[micmelesse]

@groenenboomj There is a ut with bias. see here. The github diff view is not great because the test is similar to the previous test.

[vgokhale]

@groenenboomj

@micmelesse please correct me if I'm wrong, but this isn't a "core dump issue" as such. The issue here is that triton cannot address a tensor that is > 2^31 elements. Note that it is not that Triton cannot address > 2^31 bytes. We can have several tensors in memory of size 2^30, and this will be fine. But a single tensor cannot be > 2^31.