docs: remove build container instructions from trtllm multimodal example (#4624)

Signed-off-by: Dan Gil <dagil@nvidia.com>
Signed-off-by: dagil-nvidia <dagil@nvidia.com>

docs/backends/trtllm/multinode/multinode-multimodal-example.md

@@ -19,27 +19,6 @@ limitations under the License.
 
 > **Note:** The scripts referenced in this example (such as `srun_aggregated.sh` and `srun_disaggregated.sh`) can be found in [`examples/basics/multinode/trtllm/`](https://github.com/ai-dynamo/dynamo/tree/main/examples/basics/multinode/trtllm/).
 
-> [!IMPORTANT]
-> There are some known issues in tensorrt_llm==1.1.0rc5 version for multinode multimodal support. It is important to rebuild the dynamo container with a specific version of tensorrt_llm commit to use multimodal feature.
->
-> **Build Container**
-> ```bash
-> ./container/build.sh --framework trtllm --tensorrtllm-commit b4065d8ca64a64eee9fdc64b39cb66d73d4be47c
-> ```
->
-> **Run Container**
-> ```bash
-> ./container/run.sh --framework trtllm -it
-> ```
->
-> **Update Engine Configuration Files**
->
-> Before running the deployment, you must update the engine configuration files to change `backend: DEFAULT` to `backend: default` (lowercase). Run the following command:
-> ```bash
-> sed -i 's/backend: DEFAULT/backend: default/g' /mnt/examples/backends/trtllm/engine_configs/llama4/multimodal/prefill.yaml /mnt/examples/backends/trtllm/engine_configs/llama4/multimodal/decode.yaml
-> ```
-
-
 This guide demonstrates how to deploy large multimodal models that require a multi-node setup. It builds on the general multi-node deployment process described in the main [multinode-examples.md](./multinode-examples.md) guide.
 
 Before you begin, ensure you have completed the initial environment configuration by following the **Setup** section in that guide.
@@ -186,4 +165,4 @@ pkill srun
 
 ## Known Issues
 
-- Loading `meta-llama/Llama-4-Maverick-17B-128E-Instruct` with 8 nodes of H100 with TP=16 is not posssible due to Llama4 Maverick has a config `"num_attention_heads": 40` , trtllm engine asserts on assert `self.num_heads % tp_size == 0`  causing the engine to crash on model loading.
+- Loading `meta-llama/Llama-4-Maverick-17B-128E-Instruct` with 8 nodes of H100 with TP=16 is not posssible due to Llama4 Maverick has a config `"num_attention_heads": 40` , trtllm engine asserts on assert `self.num_heads % tp_size == 0`  causing the engine to crash on model loading.
\ No newline at end of file