Add univariate vs multivariate model comparison

[Copilot]

Implements simultaneous construction and side-by-side comparison of univariate (each predictor separately) and multivariate (all predictors together) regression models, similar to autoReg package functionality.

Changes

• br_compare_models(): Builds both model types in one call, returns breg_comparison object with combined results tagged by mode
• br_show_forest_comparison(): Renders forest plot with univariate and multivariate estimates grouped by variable
• print.breg_comparison(): S3 print method for comparison objects

Usage

comparison <- br_compare_models(
  lung,
  y = c("time", "status"),
  x = c("ph.ecog", "ph.karno", "pat.karno"),
  x2 = c("age", "sex"),
  method = "coxph"
)

br_show_forest_comparison(comparison)

The forest plot displays each variable's estimates from both models, showing how effects change when adjusted for other predictors (e.g., ph.karno: HR=0.988, p=0.04 univariate → HR=1.02, p=0.11 multivariate).

Works with all model types (Cox, logistic, linear) and handles continuous/categorical variables. Added 12 tests covering various scenarios.

Original prompt

---

This section details on the original issue you should resolve

<issue_title>针对单变量和多变量模型的同时构建和对比也比较有用</issue_title>
<issue_description>这个有点 out scope，但一些东西是相通的。

https://github.com/cardiomoon/autoReg/blob/main/man/figures/unnamed-chunk-22-1.png</issue_description>

Comments on the Issue (you are @copilot in this section)

• Fixes 针对单变量和多变量模型的同时构建和对比也比较有用 #54

---

✨ Let Copilot coding agent set things up for you — coding agent works faster and does higher quality work when set up for your repo.