Add bedrock model ID fallback

[hschne]

What this does

Fix Bedrock Converse non-streaming responses so Message#model_id falls back to the request model when Bedrock omits modelId.

Why

Bedrock Converse responses do not include modelId, so non-streaming messages currently get model_id: nil. That prevents Message#cost from resolving pricing, even though the request model is known. Streaming already falls back to @model.id; this makes non-streaming match that behavior.

Reproduction

require 'ruby_llm'

RubyLLM.configure do |config|
  config.bedrock_api_key = ENV.fetch('AWS_ACCESS_KEY_ID')
  config.bedrock_secret_key = ENV.fetch('AWS_SECRET_ACCESS_KEY')
  config.bedrock_session_token = ENV['AWS_SESSION_TOKEN']
  config.bedrock_region = ENV.fetch('AWS_REGION')
end

message = RubyLLM.chat(model: 'openai.gpt-oss-120b-1:0', provider: :bedrock).ask('Say hi in one sentence.')

puts "model_id: #{message.model_id.inspect}"
puts "input_tokens: #{message.input_tokens.inspect}"
puts "output_tokens: #{message.output_tokens.inspect}"
puts "cost_total: #{message.cost.total.inspect}"

Type of change

• Bug fix
• New feature
• Breaking change
• Documentation
• Performance improvement

Scope check

• I read the Contributing Guide
• This aligns with RubyLLM's focus on LLM communication
• This isn't application-specific logic that belongs in user code
• This benefits most users, not just my specific use case

Quality check

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

AI-generated code

• I used AI tools to help write this code
• I have reviewed and understand all generated code (required if above is checked)

API changes

• Breaking change
• New public methods/classes
• Changed method signatures
• No API changes

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 83.24%. Comparing base (1080a38) to head (f205774).

Additional details and impacted files

@@            Coverage Diff             @@
##             main     #817      +/-   ##
==========================================
- Coverage   83.30%   83.24%   -0.07%     
==========================================
  Files         172      172              
  Lines        8267     8267              
  Branches     1343     1343              
==========================================
- Hits         6887     6882       -5     
- Misses        886      888       +2     
- Partials      494      497       +3     

☔ View full report in Codecov by Harness.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.