Preserves assume_model_exists in to_llm for custom models (#564)

## What this does

<!-- Clear description of what this PR does and why -->

When a user passes in the `assume_model_exists` flag through a custom
Rails model, the flag gets dropped and they are not able to utilize a
custom model.

This persists that flag or sets it to false if not provided. 

## Type of change

- [x] Bug fix
- [ ] New feature
- [ ] Breaking change
- [ ] Documentation
- [ ] Performance improvement

## Scope check

- [x] I read the [Contributing
Guide](https://github.com/crmne/ruby_llm/blob/main/CONTRIBUTING.md)
- [x] This aligns with RubyLLM's focus on **LLM communication**
- [x] This isn't application-specific logic that belongs in user code
- [x] This benefits most users, not just my specific use case

## Quality check

- [x] I ran `overcommit --install` and all hooks pass
- [x] I tested my changes thoroughly
- [ ] For provider changes: Re-recorded VCR cassettes with `bundle exec
rake vcr:record[provider_name]`
  - [x] All tests pass: `bundle exec rspec`
- [ ] I updated documentation if needed
- [x] I didn't modify auto-generated files manually (`models.json`,
`aliases.json`)

## API changes

- [ ] Breaking change
- [ ] New public methods/classes
- [ ] Changed method signatures
- [x] No API changes

## Related issues

<!-- Link issues: "Fixes #123" or "Related to #123" -->
#555

Co-authored-by: Carmine Paolino <carmine@paolino.me>

Author: creaumond

lib/ruby_llm/active_record/chat_methods.rb

@@ -79,7 +79,8 @@ def to_llm
         model_record = model_association
         @chat ||= (context || RubyLLM).chat(
           model: model_record.model_id,
-          provider: model_record.provider.to_sym
+          provider: model_record.provider.to_sym,
+          assume_model_exists: assume_model_exists || false
         )
         @chat.reset_messages!
 

spec/ruby_llm/active_record/acts_as_model_spec.rb

@@ -255,7 +255,8 @@ def messages
         # Mock the chat creation to verify parameters
         expect(RubyLLM).to receive(:chat).with( # rubocop:disable RSpec/MessageSpies,RSpec/StubbedMock
           model: 'test-gpt',
-          provider: :openai
+          provider: :openai,
+          assume_model_exists: false
         ).and_return(
           instance_double(RubyLLM::Chat, reset_messages!: nil, add_message: nil,
                                          instance_variable_get: {}, on_new_message: nil, on_end_message: nil,
@@ -272,7 +273,8 @@ def messages
 
         expect(RubyLLM).to receive(:chat).with( # rubocop:disable RSpec/MessageSpies,RSpec/StubbedMock
           model: 'test-claude',
-          provider: :anthropic
+          provider: :anthropic,
+          assume_model_exists: false
         ).and_return(
           instance_double(RubyLLM::Chat, reset_messages!: nil, add_message: nil,
                                          instance_variable_get: {}, on_new_message: nil, on_end_message: nil,

spec/ruby_llm/active_record/acts_as_spec.rb

@@ -535,6 +535,14 @@ class ToolCall < ActiveRecord::Base # rubocop:disable RSpec/LeakyConstantDeclara
         expect(llm_tool_call.name).to eq('calculator')
         expect(llm_tool_call.arguments).to eq({ 'expression' => '2 + 2' })
       end
+
+      it 'correctly preserves custom model' do
+        custom_model = 'my-custom-model'
+        bot_chat = Assistants::BotChat.create!(model: custom_model, assume_model_exists: true, provider: 'openrouter')
+        bot_chat.save!
+        llm_chat = bot_chat.to_llm
+        expect(llm_chat.model.id).to eq(custom_model)
+      end
     end
   end