docs: clarify CSV non-adjacent column limitation

Author: BohuTANG

docs/cn/guides/20-self-hosted/04-references/http-streaming-load.md

@@ -84,6 +84,8 @@ X-Databend-SQL: insert into demo.people(name,age,city) values (?, ?, 'BJ') from
     ```text
     X-Databend-SQL: insert into demo.people(id,name) from @_databend_load file_format=(type=csv skip_header=1 error_on_column_count_mismatch=false)
     ```
+  - 这个能力只适用于“取前 N 列”的场景。Streaming load 按字段位置映射，不支持挑选非连续列（例如 `id,name,age` 想只导入 `id` 和 `age`）。
+    - 解决思路：先在本地把 CSV 预处理成只包含需要的列，或先上传到 stage 再用 `SELECT $1, $3 FROM @stage/file.csv` 这种方式做列投影。
 
 **cURL 模板：**
 

docs/en/guides/20-self-hosted/04-references/http-streaming-load.md

@@ -84,6 +84,8 @@ X-Databend-SQL: insert into demo.people(name,age,city) values (?, ?, 'BJ') from
     ```text
     X-Databend-SQL: insert into demo.people(id,name) from @_databend_load file_format=(type=csv skip_header=1 error_on_column_count_mismatch=false)
     ```
+  - This only helps when you want the **first N fields**. Streaming load maps CSV fields by position and does not support selecting non-adjacent fields (for example, `id,name,age` → insert only `id` and `age`).
+    - Workaround: preprocess the CSV to keep only the needed columns, or load via stage and project columns (for example, `SELECT $1, $3 FROM @stage/file.csv`).
 
 **cURL template:**