Fix swa available memory check

[ispobock]

Motivation

When running swa model like mimo-v2-flash for a while, it will raise error:

[2025-12-26 03:42:15 DP1 TP4] Scheduler hit an exception: Traceback (most recent call last):
  File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 2932, in run_scheduler_process
    scheduler.event_loop_overlap()
  File "/usr/local/lib/python3.12/dist-packages/torch/utils/_contextlib.py", line 120, in decorate_context
    return func(*args, **kwargs)
           ^^^^^^^^^^^^^^^^^^^^^
  File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 1122, in event_loop_overlap
    batch = self.get_next_batch_to_run()
            ^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 1844, in get_next_batch_to_run
    self.running_batch = self.update_running_batch(self.running_batch)
                         ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/sgl-workspace/sglang/python/sglang/srt/managers/scheduler.py", line 2093, in update_running_batch
    old_available_tokens = self.token_to_kv_pool_allocator.available_size()
                           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/sgl-workspace/sglang/python/sglang/srt/mem_cache/allocator.py", line 215, in available_size
    raise NotImplementedError()
NotImplementedError

It's related to this change: #15694
The available size of swa token_to_kv_pool_allocator should get seperately, ref changes in #7367

cc: @hnyls2002 @hanming-lu for double-check

[gemini-code-assist]

Warning

You have reached your daily quota limit. Please wait up to 24 hours and I will start processing your requests again!

[ispobock]

/tag-and-rerun-ci