fix: force milvus-lite installation for inline::milvus (#3488)

# What does this PR do?

pymilvus recently made `milvus-lite` an optional dependency to their
package. If someone wants to use the inline provider we must include the
extra dependency.
For more details see: milvus-io/pymilvus#2976

Signed-off-by: Sébastien Han <[email protected]>

Author: leseb

docs/source/providers/vector_io/remote_milvus.md

@@ -23,7 +23,13 @@ To use Milvus in your Llama Stack project, follow these steps:
 
 ## Installation
 
-You can install Milvus using pymilvus:
+If you want to use inline Milvus, you can install:
+
+```bash
+pip install pymilvus[milvus-lite]
+```
+
+If you want to use remote Milvus, you can install:
 
 ```bash
 pip install pymilvus

llama_stack/providers/registry/vector_io.py

@@ -633,7 +633,13 @@ def available_providers() -> list[ProviderSpec]:
 
 ## Installation
 
-You can install Milvus using pymilvus:
+If you want to use inline Milvus, you can install:
+
+```bash
+pip install pymilvus[milvus-lite]
+```
+
+If you want to use remote Milvus, you can install:
 
 ```bash
 pip install pymilvus
@@ -807,7 +813,7 @@ def available_providers() -> list[ProviderSpec]:
         InlineProviderSpec(
             api=Api.vector_io,
             provider_type="inline::milvus",
-            pip_packages=["pymilvus>=2.4.10"],
+            pip_packages=["pymilvus[milvus-lite]>=2.4.10"],
             module="llama_stack.providers.inline.vector_io.milvus",
             config_class="llama_stack.providers.inline.vector_io.milvus.MilvusVectorIOConfig",
             api_dependencies=[Api.inference],