Encoded message too large

**Describe the bug**

```
{ │
│ runner "query_id": "TPCH-21", │
│ runner "has_errors": true, │
│ runner "has_warnings": false, │
│ runner "error": "DataFusion error: Execution error: Status { code: OutOfRange, message: \"Error, encoded message length too large: found 19282335 bytes, the limit is: 16777216 bytes\", metadata: MetadataMap { headers: {" │
│ runner }, │
│ runner { │
│ runner "query_id": "TPCH-22", │
│ runner "has_errors": true, │
│ runner "has_warnings": false, │
│ runner "error": "DataFusion error: Execution error: Status { code: OutOfRange, message: \"Error, encoded message length too large: found 23232616 bytes, the limit is: 16777216 bytes\", metadata: MetadataMap { headers: {" │
│ runner }
```

**To Reproduce**
Steps to reproduce the behavior:

**Expected behavior**
A clear and concise description of what you expected to happen.

**Additional context**
Add any other context about the problem here.

Start by reproducing the TPCH-21 and TPCH-22 cases and tracing where the DataFusion execution error is returned. Identify which component imposes the 16,777,216-byte encoded-message limit; done means both queries no longer fail with the reported encoded message size error.

Written by the indexing model from the issue text.