Aggregate queries using group by doesn't return the correct number of rows

### Describe the bug
When running a query with group by and built-in functions in Clickhouse the correct result is not returned compared with running the exact same query on the Clickhouse command line. This was tested within actix-web.

### Steps to reproduce
1. Run query on the Clickhouse CLI.
2. Execute database query within route handler in actix-web

### Expected behaviour
Only 1 - 3 rows are returned when the limit number should be returned.

### Code example
```
use actix_web::{get, web, HttpResponse, Responder};
use clickhouse::{Client, Row};
use serde::{Deserialize, Serialize};
use time::{OffsetDateTime};

#[derive(Debug, Row, Deserialize, Serialize)]
struct StockRow<'a> {
ticker: &'a str,
#[serde(with = "clickhouse::serde::time::datetime")]
stock_timestamp: OffsetDateTime,
open: f32,
high: f32,
low: f32,
close: f32,
volume: u32,
}

#[get("/api/data")]
async fn api_data(
client: web::Data,
params: web::Query,
) -> Result {
let ticker = ¶ms.ticker;
let limit: i16 = match params.limit {
Some(limit) => limit,
None => 100,
};

let mut cursor = client
.query("SELECT ticker, toStartOfMinute(stock_timestamp) stock_timestamp, argMin(open, stock_timestamp) AS open,
MAX(high) AS high,
MIN(low) AS low,
argMax(close, stock_timestamp) AS close,
SUM(volume) AS volume FROM stocks WHERE ticker = ? GROUP BY ticker, stock_timestamp ORDER BY stock_timestamp LIMIT ?")
.bind(ticker)
.bind(limit)
.fetch::>()
.unwrap();

while let Ok(Some(row)) = cursor.next().await {
println!("{:?}", row);
}

Ok(HttpResponse::Ok().body(""))
}
```

### Error log
No errors, incorrect result is returned.

### Query log
This query can be tested using the Clickhouse CLI:

```
SELECT
ticker,
toStartOfMinute(stock_timestamp) AS stock_timestamp,
argMin(open, stock_timestamp) AS open,
MAX(high) AS high,
MIN(low) AS low,
argMax(close, stock_timestamp) AS close,
SUM(volume) AS volume
FROM stocks
WHERE ticker = 'NVDA'
GROUP BY ticker, stock_timestamp
ORDER BY stock_timestamp
LIMIT 100;
```

### Configuration
#### Environment
* Client version: 0.13.1 (Clickhouse client CLI: 25.1.5.31)
* OS: Ubuntu 24.04.2 LTS

#### ClickHouse server
* ClickHouse Server version: 25.1.2
* ClickHouse Server non-default settings, if any:
* `CREATE TABLE` statements for tables involved:
```
CREATE TABLE IF NOT EXISTS test_data.stocks (
ticker LowCardinality(String),
stock_timestamp DateTime CODEC(Delta, ZSTD),
open Float32 CODEC(ZSTD),
high Float32 CODEC(ZSTD),
low Float32 CODEC(ZSTD),
close Float32 CODEC(ZSTD),
volume UInt32 CODEC(ZSTD),
timestamp DateTime CODEC(Delta, ZSTD)
) ENGINE = MergeTree()
PARTITION BY toYYYYMM(stock_timestamp)
ORDER BY (ticker, stock_timestamp);
```

* Sample data for all these tables, use [clickhouse-obfuscator](https://github.com/ClickHouse/ClickHouse/blob/master/programs/obfuscator/Obfuscator.cpp#L42-L80) if necessary

Sample data was outputted as JSON from Clickhouse:

```
{
"row_1": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:00:00","open":124.91,"high":124.93,"low":124.3,"close":124.5,"volume":20040123,"timestamp":"2025-02-27 14:30:34"},
"row_2": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:00:10","open":124.52,"high":125.06,"low":124.36,"close":124.42,"volume":157265,"timestamp":"2025-02-27 14:30:34"},
"row_3": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:00:20","open":124.42,"high":124.85,"low":124.3,"close":124.44,"volume":46094,"timestamp":"2025-02-27 14:30:34"},
"row_4": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:00:30","open":124.53,"high":124.62,"low":124.37,"close":124.41,"volume":19278,"timestamp":"2025-02-27 14:30:34"},
"row_5": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:00:40","open":124.49,"high":124.56,"low":124.4,"close":124.4,"volume":54513,"timestamp":"2025-02-27 14:30:34"},
"row_6": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:00:50","open":124.42,"high":124.51,"low":124.38,"close":124.51,"volume":19510,"timestamp":"2025-02-27 14:30:34"},
"row_7": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:01:00","open":124.52,"high":124.57,"low":124.41,"close":124.43,"volume":33862,"timestamp":"2025-02-27 14:30:34"},
"row_8": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:01:10","open":124.48,"high":124.5,"low":124.25,"close":124.4,"volume":44072,"timestamp":"2025-02-27 14:30:34"},
"row_9": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:01:20","open":124.41,"high":124.41,"low":124.28,"close":124.36,"volume":21589,"timestamp":"2025-02-27 14:30:34"},
"row_10": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:01:30","open":124.35,"high":124.43,"low":124.29,"close":124.29,"volume":46371,"timestamp":"2025-02-27 14:30:34"},
"row_11": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:01:40","open":124.29,"high":124.47,"low":124.29,"close":124.42,"volume":58016,"timestamp":"2025-02-27 14:30:34"},
"row_12": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:01:50","open":124.43,"high":124.5,"low":124.42,"close":124.5,"volume":13858,"timestamp":"2025-02-27 14:30:34"},
"row_13": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:02:00","open":124.49,"high":124.71,"low":124.49,"close":124.71,"volume":34373,"timestamp":"2025-02-27 14:30:34"},
"row_14": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:02:10","open":124.68,"high":124.75,"low":124.64,"close":124.7,"volume":53706,"timestamp":"2025-02-27 14:30:34"},
"row_15": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:02:20","open":124.72,"high":124.79,"low":124.7,"close":124.79,"volume":8585,"timestamp":"2025-02-27 14:30:34"},
"row_16": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:02:30","open":124.75,"high":124.85,"low":124.74,"close":124.85,"volume":27377,"timestamp":"2025-02-27 14:30:34"},
"row_17": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:02:40","open":124.84,"high":124.91,"low":124.83,"close":124.88,"volume":47918,"timestamp":"2025-02-27 14:30:34"},
"row_18": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:02:50","open":124.83,"high":124.9,"low":124.77,"close":124.83,"volume":37207,"timestamp":"2025-02-27 14:30:34"},
"row_19": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:03:00","open":124.84,"high":124.95,"low":124.83,"close":124.94,"volume":57767,"timestamp":"2025-02-27 14:30:34"},
"row_20": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:03:10","open":124.93,"high":124.97,"low":124.65,"close":124.95,"volume":459171,"timestamp":"2025-02-27 14:30:34"},
"row_21": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:03:20","open":124.91,"high":124.93,"low":124.91,"close":124.92,"volume":23759,"timestamp":"2025-02-27 14:30:34"},
"row_22": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:03:30","open":124.91,"high":124.93,"low":124.91,"close":124.91,"volume":19199,"timestamp":"2025-02-27 14:30:34"},
"row_23": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:03:40","open":124.9,"high":124.93,"low":124.83,"close":124.83,"volume":58486,"timestamp":"2025-02-27 14:30:34"},
"row_24": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:03:50","open":124.83,"high":124.85,"low":124.8,"close":124.84,"volume":8143,"timestamp":"2025-02-27 14:30:34"},
"row_25": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:04:00","open":124.83,"high":124.84,"low":124.81,"close":124.83,"volume":19816,"timestamp":"2025-02-27 14:30:34"},
"row_26": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:04:10","open":124.83,"high":124.84,"low":124.82,"close":124.83,"volume":14417,"timestamp":"2025-02-27 14:30:34"},
"row_27": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:04:20","open":124.84,"high":124.84,"low":124.7,"close":124.7,"volume":18907,"timestamp":"2025-02-27 14:30:34"},
"row_28": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:04:30","open":124.7,"high":124.87,"low":124.7,"close":124.85,"volume":34307,"timestamp":"2025-02-27 14:30:34"},
"row_29": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:04:40","open":124.8,"high":124.92,"low":124.65,"close":124.9,"volume":7008442,"timestamp":"2025-02-27 14:30:34"},
"row_30": {"ticker":"NVDA","stock_timestamp":"2025-01-30 21:04:50","open":124.86,"high":124.93,"low":124.84,"close":124.89,"volume":9378,"timestamp":"2025-02-27 14:30:34"}
}
```

Start by running the supplied aggregate query through the ClickHouse CLI and then through the Rust client code in the actix-web route, comparing the returned row counts and values. Inspect the client.query, bind, fetch, and cursor.next flow, especially the LIMIT binding and grouped timestamp expression. Done means the Rust client returns the same grouped rows as the CLI.

Written by the indexing model from the issue text.