Claude review: arm64: dts: qcom: sdm670-google: add common device tree include

[Claude Code Review Bot <claude-review@example.com>]

Patch Review

**Status: Good refactoring, one concern.**

The factoring approach is correct. The common dtsi removes device-specific properties (`model`, `compatible`, `charge-full-design-microamp-hours`, `height`, `touchscreen-x-mm`, `touchscreen-y-mm`, and panel `compatible`) and adds labels (`framebuffer:`, `fb_mem:`, `panel:`, `rmi4_f12:`) for overriding in device-specific files.

1. **Framebuffer `height` comment**: The common dtsi replaces the `height` property with a comment:
   ```
   +			/* height is in specific device trees */
   ```
   This is fine for documentation, but it means the common framebuffer node lacks a required `height` property for `simple-framebuffer`. If used standalone (which it isn't, as a `.dtsi`), this would be invalid. This is acceptable since the `.dtsi` is always included by a `.dts` that provides `height`, but the DT compiler may not warn if a `.dts` forgets to set it.

2. **Memory-region change**: The framebuffer changes from `reg` to `memory-region`:
   ```
   -			reg = <0 0x9c000000 0 (1080 * 2220 * 4)>;
   +			memory-region = <&fb_mem>;
   ```
   This is a good change since it references the reserved memory region properly, but it changes the binding usage for the existing sargo board. The `fb_mem` region is `0x2400000` bytes (about 37.7 MB) which is larger than `1080 * 2220 * 4 = 9,590,400` bytes. This should still work since `simple-framebuffer` only reads the needed portion, but the semantics are slightly different.

3. **License change**: The SPDX change from `GPL-2.0` to `GPL-2.0-only` is well-explained in the commit message.

---
Generated by Claude Code Patch Reviewer