feat: Add continuous outcomes support for MAIC analysis

[choxos]

This commit implements comprehensive support for continuous outcomes in MAIC (Matching-Adjusted Indirect Comparison) analyses, including both anchored and unanchored comparisons.

New Features:

• Effect measures: Mean Difference (MD), Standardized Mean Difference (SMD), and Ratio of Means (RoM)
• SMD calculation methods: Cohen's d, Hedges' g, and Glass's delta
• Helper function get_pseudo_ipd_continuous() to generate pseudo IPD from aggregated continuous outcome data
• Helper function lm_makeup() to summarize linear model results
• Support for robust standard errors using HC3 estimator
• Integration with existing maic_anchored() and maic_unanchored() functions

New Datasets:

• adlb_sat: Single-arm continuous outcome data (500 subjects)
• adlb_twt: Two-arm continuous outcome data (1000 subjects)

Documentation:

• Comprehensive vignettes for anchored and unanchored continuous analysis
• Full documentation for all new functions and datasets
• Example scripts demonstrating usage

Testing:

• 33 new tests covering continuous helper functions and MAIC analyses
• All tests pass (174/174 total)
• R CMD check passes with no errors or warnings
• Integration tests confirm compatibility with existing binary and TTE endpoints

Technical Details:

• Weighted linear regression using lm() with MAIC weights
• Robust covariance estimation via sandwich::vcovHC()
• Proper handling of edge cases (zero SD, positive values for RoM)
• Consistent USUBJID mapping between outcome and baseline datasets