EXPERIMENTAL_tx_status will give incomplete response after we release resharding

[telezhnaya]

Underneath, it uses get_tx_status with fetch_receipt option.

We check if we are tracking current shard

nearcore/chain/client/src/view_client.rs

Line 439 in 06aebc3

// Check if we are tracking this shard.

If not, we add the request to the queue

nearcore/chain/client/src/view_client.rs

Line 473 in 06aebc3

if Self::need_request(tx_hash, &mut request_manager.tx_status_requests) {

Then, if the request is repeated, we should get it from the cache

nearcore/chain/client/src/view_client.rs

Lines 426 to 432 in 06aebc3

	{
	let mut request_manager = self.request_manager.write().expect(POISONED_LOCK_ERR);
	if let Some(res) = request_manager.tx_status_response.pop(&tx_hash) {
	request_manager.tx_status_requests.pop(&tx_hash);
	return Ok(Some(FinalExecutionOutcomeViewEnum::FinalExecutionOutcome(res)));
	}
	}

And we ignore fetch_receipt flag there.

It's not reproducible now because we haven't released resharding and we just don't use this logic. After introducing the release, we will not show requested receipts in the response.