[ModelRunner] Use shared CachedRequestData cross request to fix ci (#1546)

### What this PR does / why we need it?

This PR (adapted from
2863befce3)
updates the CachedRequestData definition to use a single instance shared
across all requests in a batch, instead of creating a new instance per
request.

Found ci boken by the vllm's model_runner change: `ERROR 07-01 09:53:53
[core.py:521] TypeError: 'CachedRequestData' object is not iterable`,
Modify the model_runner to fix it.


### Does this PR introduce _any_ user-facing change?
No

### How was this patch tested?
pass ci will verify this.

---------

Signed-off-by: ganyi <pleaplusone.gy@gmail.com>
Signed-off-by: Yikun Jiang <yikunkero@gmail.com>
Co-authored-by: Yikun Jiang <yikunkero@gmail.com>

tests/e2e/singlecard/test_scheduler.py

@@ -192,7 +192,10 @@ def test_schedule(enable_prefix_caching: Optional[bool],
     # Test initial scheduling
     output = scheduler.schedule()
     assert len(output.scheduled_new_reqs) == len(requests)
-    assert len(output.scheduled_cached_reqs) == 0
+    if vllm_version_is("0.9.1"):
+        assert len(output.scheduled_cached_reqs) == 0
+    else:
+        assert output.scheduled_cached_reqs.num_reqs == 0
     assert len(output.finished_req_ids) == 0
     # Verify all requests are scheduled.
     for req_id, num_tokens in output.num_scheduled_tokens.items():