add support for custom llm with ssl support (#380)

* add support for custom llm with ssl support

Add support for using custom llm that are served through https protocol.

* add instructions on how to add custom inference endpoint

* fix formatting

* add more details

* Apply suggestions from code review

Co-authored-by: Salman Paracha <salman.paracha@gmail.com>

* Apply suggestions from code review

* fix precommit

---------

Co-authored-by: Salman Paracha <salman.paracha@gmail.com>

arch/arch_config_schema.yaml

@@ -61,6 +61,11 @@ properties:
           type: boolean
         endpoint:
           type: string
+        protocol:
+          type: string
+          enum:
+            - http
+            - https
       additionalProperties: false
       required:
         - name

arch/envoy.template.yaml

@@ -567,6 +567,18 @@ static_resources:
                       address: {{ local_llm_provider.endpoint }}
                       port_value: {{ local_llm_provider.port }}
                   hostname: {{ local_llm_provider.endpoint }}
+      {% if local_llm_provider.protocol == "https" %}
+      transport_socket:
+        name: envoy.transport_sockets.tls
+        typed_config:
+          "@type": type.googleapis.com/envoy.extensions.transport_sockets.tls.v3.UpstreamTlsContext
+          sni: {{ local_llm_provider.endpoint }}
+          common_tls_context:
+            tls_params:
+              tls_minimum_protocol_version: TLSv1_2
+              tls_maximum_protocol_version: TLSv1_3
+      {% endif %}
+
 {% endfor %}
     - name: arch_internal
       connect_timeout: 5s