`get_sequential_data` returns disordered results

Required prerequisites

• Consult the security policy. If reporting a security vulnerability, do not report the bug using this form. Use the process described in the policy to report the issue.
• Make sure you've read the documentation. Your issue may be addressed there.
• Search the issue tracker to verify that this hasn't already been reported. +1 or comment there if it has.
• If possible, make a PR with a failing test to give us a starting point to work on!

Describe the bug

When using registers to get multiple measurements from a shot, if one uses get_sequential_data to process the results, the positions of measurement outcomes are not consistent with the shot index. This makes it hard to group different measurements of a single shot together. For example, when one runs

cudaq_counts = cudaq.sample(test_kernel,  shots_count=3)
count_lists = [cudaq_counts.get_sequential_data(f'r_{r}') for r in range(3)]

The ideal results should be

[
[results_for_shot1_r1, results_for_shot2_r1, results_for_shot3_r1],
[results_for_shot1_r2, results_for_shot2_r2, results_for_shot3_r2],
[results_for_shot1_r3, results_for_shot2_r3, results_for_shot3_r3],
]

But instead, one can get

[
[results_for_shot1_r1, results_for_shot2_r1, results_for_shot3_r1],
[results_for_shot1_r2, results_for_shot2_r2, results_for_shot3_r2],
[results_for_shot3_r3, results_for_shot1_r3, results_for_shot2_r3],
]

Steps to reproduce the bug

import cudaq

k = cudaq.make_kernel()
q = k.qalloc(2)
k.h(q)
k.mz(q[0], 'r_0')
k.x(q[0])
k.mz(q, 'r_1')

counts = cudaq.sample(k, shots_count=5)
count_lists = [counts.get_sequential_data(f'r_{r}') for r in range(2)]
print(count_lists)

Expected behavior

The 1st measurement result (r_0) is random on q[0]. Then the 2nd measurement (r_1) should be deterministic on q[0] but random on q[1]. But actually I got

[['1', '1', '0', '0', '0'], ['11', '11', '00', '00', '10']]

The two shots where q[0] measured to be '1' in r_0 should give '0x' in r_1, and the position are mismatched (index 0,1 and index 2,3)

Is this a regression? If it is, put the last known working version (or commit) here.

Not a regression

Environment

• CUDA-Q version: amd64-cu12-0.12.0
• Python version: 3.10.12
• Operating system: Linux(from the official docker image ghcr.io/nvidia/cudaqx:latest)

Suggestions

No response

Reproduce the issue with the provided Python kernel and the get_sequential_data calls. Start by tracing get_sequential_data and compare the returned lists with the shot order shown in the expected behavior. Done means measurements from each register remain aligned by shot index across sequential results.