Accept memory_format kwarg on rand/randn/randint _like ops

[fruet (gabrielfruet)]

Exporting rand_like, randn_like, or randint_like with memory_format=torch.preserve_format raised a ConversionError: the torchlib signatures dropped the memory_format kwarg the aten schema declares. Adds it as a no-op trailing kwarg on the four *_like random ops, matching what #2994 did for full_like.

Fixes #3002.

Changes:

• ops/core.py: add memory_format to aten_rand_like, aten_randn_like, aten_randint_like, aten_randint_like_low_dtype
• e2e tests: export each op with memory_format=torch.preserve_format

Testing: the new tests assert export succeeds rather than value equality, since these ops are non-deterministic.

[Justin Chu (justinchuby)]

Could you resolve merge conflicts? Thanks

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 72.63%. Comparing base (9ef0aac) to head (0417850).
⚠️ Report is 14 commits behind head on main.

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #3011   +/-   ##
=======================================
  Coverage   72.63%   72.63%           
=======================================
  Files         265      265           
  Lines       32204    32210    +6     
  Branches     3041     3042    +1     
=======================================
+ Hits        23391    23397    +6     
  Misses       7779     7779           
  Partials     1034     1034           

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

[fruet (gabrielfruet)]

Justin Chu (@justinchuby) Conflicts are resolved! Thank you for approving :)

It still needs a re-approval since I pushed again