docs(lora): add YOLO-Master domain adaptation configs

[xie-guangzhen]

Description | 描述

Resubmission after the valid RhinoBird claim window opened on July 1.

Adds YOLO-Master-EsMoE-N LoRA adaptation examples for two visually different vertical scenarios requested in #50:

• VisDrone dense aerial detection
• brain-tumor sparse medical detection

The PR includes scenario-specific LoRA configs, a reproducible rank sweep launcher, and README guidance for rank selection, target module policy, MoE routing handling, and common pitfalls.

Related Issue | 关联 Issue

Fixes #50

Files Added | 新增文件

• examples/lora_examples/yolo_master_visdrone_lora.yaml
• examples/lora_examples/yolo_master_brain_tumor_lora.yaml
• examples/lora_examples/domain_lora_rank_sweep.py

Self-test Checklist | 自测清单

• Python syntax check passed for the sweep script
• VisDrone dry-run emits expected r=4,8,16 commands
• brain-tumor dry-run emits expected r=4,8,16 commands
• git diff --check passed

Validation commands run on the original implementation branch:

python3 -m py_compile examples/lora_examples/domain_lora_rank_sweep.py
python3 examples/lora_examples/domain_lora_rank_sweep.py --scenario visdrone --dry-run
python3 examples/lora_examples/domain_lora_rank_sweep.py --scenario brain_tumor --dry-run
git diff --check