Avoid a C-order copy in dpnp.einsum

[vlad-perevezentsev]

This PR improves dpnp.einsum performance by avoiding a copy into C-order.

Commit 039b7f1 (#3058) remaps order="K" to "C" which turned the final dpnp.asarray from a no-op into a full device copy because dpnp contracts operands in reverse order so "ij,jk" computes (a @ b).T and the result is naturally f-contiguous

This PR builds the result directly in the requested layout instead of fixing it afterwards. The last contraction determines the output axis order, so it can choose which operand is on the left side of the matmul and avoid an extra copy.
This is only applied for "C" order, since for other layouts the default operand order already matches the expected order="K" + optimize=True behavior. The swap does not move any data. It only changes the operand order in the matmul

The performance result bench_linalg.MatMul.time_einsum_ij_jk on PVC:

dtype	size	main	fix
float64	1024	0.953 ms	0.612 ms (x0.64)
float64	4096	8.590 ms	7.490 ms (x0.87)
float32	1024	0.574 ms	0.439 ms (x0.76)
float32	4096	7.577 ms	7.005 ms (x0.92)
int32	1024	1.891 ms	1.301 ms (x0.69)
int32	4096	54.324 ms	35.045 ms (x0.65)
int64	1024	4.637 ms	3.735 ms (x0.81)
int64	4096	200.454 ms	162.920 ms (x0.81)

• Have you provided a meaningful PR description?
• Have you added a test, reproducer or referred to an issue with a reproducer?
• Have you tested your changes locally for CPU and GPU devices?
• Have you made sure that new changes do not introduce compiler warnings?
• Have you checked performance impact of proposed changes?
• Have you added documentation for your changes, if necessary?
• Have you added your changes to the changelog?

[github-actions]

View rendered docs @ https://intelpython.github.io/dpnp/pull/3069/index.html

[github-actions]

Array API standard conformance tests for dpnp=0.21.0dev8=py314ha0e2e8e_11 ran successfully.
Passed: 1376
Failed: 0
Skipped: 6

[coveralls]

coverage: 78.48% (+0.003%) from 78.477% — fix_perf_einsum into master

[ndgrigorian]

LGTM