Please make it possible to "mark" subtests as `xfail`

What's the problem this feature will solve?

Currently, it isn't possible to conveniently mark a subtest as XFAIL-ing with full support for xfail_strict=True.

Describe the solution you'd like

Ideally, I'd like to be able to do something like:

def test_foo(subtests):
    with subtests.test(msg=..., xfail=condition):
        ...

that would mark this one subtest as XFAIL, the same way pytest.mark.xfail marks the whole test as XFAIL-ing.

Alternative Solutions

1. Marking the whole test with pytest.mark.xfail causes it to unconditionally fail with xfail_strict=True. Reading the comments in the code, I infer that it isn't supported. And even if it were, more precision would be desirable.

   $ pytest test.py -v -o xfail_strict=True
   ========================================================= test session starts =========================================================
   platform linux -- Python 3.12.12, pytest-9.1.0.dev148+g0e9db5fa5, pluggy-1.6.0 -- /tmp/foo/.venv/bin/python3
   cachedir: .pytest_cache
   rootdir: /tmp/foo
   collected 1 item                                                                                                                      
   
   test.py::test_foo SUBXFAIL[xfail] (foo)                                                                                         [100%]
   test.py::test_foo FAILED                                                                                                        [100%]
   
   ============================================================== FAILURES ===============================================================
   ______________________________________________________________ test_foo _______________________________________________________________
   [XPASS(strict)] foo
   ======================================================= short test summary info =======================================================
   FAILED test.py::test_foo - [XPASS(strict)] foo
   ==================================================== 1 failed, 1 xfailed in 0.01s =====================================================
   

2. It is possible to call pytest.xfail() directly from the test. However, this requires a bit of handiwork to "reimplement" xfail_strict.

Additional context

n/a

Start by tracing how subtests.test, pytest.mark.xfail, and pytest.xfail interact with xfail_strict=True. Compare the existing subtest reporting and strict-XFAIL behavior, then verify that the proposed per-subtest xfail condition reports expected failures and strict unexpected passes without affecting other subtests.