refactor!: improve provider resolution

Improve the way how different providers are resolved and configured. Make sure
the output and input processing is more consistent between providers by
providing common interface that providers must implement.

Improvements:
- Change embeddings field to embed for clarity
- Add proper types for provider inputs and outputs
- Make provider resolution more generic
- Make common provider interface clearer
- Add types for model and agent configs
- Make prepare_input and prepare_output more consistent

Signed-off-by: Tomas Slusny <slusnucky@gmail.com>

Author: deathbeam

README.md

@@ -397,7 +397,11 @@ Here's how to implement an [ollama](https://ollama.com/) provider:
 {
   providers = {
     ollama = {
-      embeddings = 'copilot_embeddings', -- Use Copilot as embedding provider
+      embed = 'copilot_embeddings', -- Use Copilot as embedding provider
+
+      -- Copy copilot input and output processing
+      prepare_input = require('CopilotChat.config.providers').copilot.prepare_input,
+      prepare_output = require('CopilotChat.config.providers').copilot.prepare_output,
 
       get_headers = function()
         return {
@@ -422,14 +426,6 @@ Here's how to implement an [ollama](https://ollama.com/) provider:
         return models
       end,
 
-      prepare_input = function(inputs, opts)
-        return {
-          model = opts.model,
-          messages = inputs,
-          stream = true,
-        }
-      end,
-
       get_url = function()
         return 'http://localhost:11434/api/chat'
       end,