refactor: backend/service split, channel-based llm flow (#1963)

Refactor: channel based llm flow and services split

---------

Signed-off-by: Dave Lee <dave@gray101.com>

examples/bruno/LocalAI Test Requests/llm text/-completions Stream.bru

@@ -0,0 +1,25 @@
+meta {
+  name: -completions Stream
+  type: http
+  seq: 4
+}
+
+post {
+  url: {{PROTOCOL}}{{HOST}}:{{PORT}}/completions
+  body: json
+  auth: none
+}
+
+headers {
+  Content-Type: application/json
+}
+
+body:json {
+  {
+      "model": "{{DEFAULT_MODEL}}",
+      "prompt": "function downloadFile(string url, string outputPath) {",
+      "max_tokens": 256,
+      "temperature": 0.5,
+      "stream": true
+  }
+}