Fix measured size followup review regressions

[shiny-code-bot]

Summary

• Split measured size-budget alignment so over-target samples can move smaller while under-target samples can spend bitrate on quality.
• Narrow measured-size follow-up detection to notes with explicit post-sample or smaller-revision context.
• Add regression coverage for plain size sample notes and under-target quality increases.

Verification

• uv run --with pytest pytest -q tests/test_tuning_runtime.py -k "plain_size_sample_as_followup or measured_size_followup or quality_increase or quality_tradeoff or under_target_size_budget_quality"
• uv run --with pytest pytest -q tests/test_tuning_runtime.py
• bash scripts/pre-commit-check.sh
• uv run /Users/cbusillo/Developer/codex-skills/jetbrains-inspection/scripts/jb-inspect.py closeout --repo /Users/cbusillo/Developer/mediaforce --scope changed_files