[Test][Misc] Move 2 size cases to double_node (#15141)

### What this PR does / why we need it?
This PR updates the test configuration for
`DeepSeek-V4-Pro-w4a8-PD-no-prefix` by reducing the
`--max-num-batched-tokens` from `8192` to `4096` to optimize memory
usage or adjust test sizes. Additionally, it introduces a new multi-node
end-to-end test configuration for `GLM-5.2-W8A8C8-A3-Memcache-PD` with
disaggregated prefill routing and memcache KV pool.
### Does this PR introduce _any_ user-facing change?
No
### How was this patch tested?
By running weekly test.
- vLLM version: v0.27.1
- vLLM main:
https://github.com/vllm-project/vllm/commit/ba07e4a48fc951300d97eb506217dd530583dea3

---------

Signed-off-by: chen-commits <1636718796@qq.com>
Signed-off-by: chen <1636718796@qq.com>