Pass --model CLI override to create_llm_client log line

[daniel-thom]

Summary

The ask, generate, and verify commands all accept a --model flag that resolve_settings returns as resolved_model. Each command then calls create_llm_client with settings.llm.model (the .env value) instead of resolved_model, so the "Using {provider} LLM backend (model=…)" log line in llm.py:952 shows the .env model rather than the override the user actually requested.

This was logging-only — the actual API calls in run_agent_loop already use resolved_model correctly (cli.py:231, generate.py:411, verify.py:118/125) — but the misleading log line made it look like the override wasn't being applied.

Files changed

• src/datasight/cli.py:133 — run_ask_pipeline
• src/datasight/cli_commands/generate.py:312
• src/datasight/cli_commands/verify.py:100

Test plan

• prek run --all-files passes
• pytest -m \"not integration\" passes (1481 tests)
• Manual verification: datasight ask --model=X --file q.txt shows model=X in the startup log

🤖 Generated with Claude Code

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 86.90%. Comparing base (3282c3b) to head (dffe174).

Additional details and impacted files

@@           Coverage Diff           @@
##             main      #69   +/-   ##
=======================================
  Coverage   86.90%   86.90%           
=======================================
  Files          61       61           
  Lines       11462    11462           
=======================================
  Hits         9961     9961           
  Misses       1501     1501           

☔ View full report in Codecov by Sentry.
📢 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.
• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.