Add preliminary QCDL support

[randomir]

Example:

import orjson
from dwave.cloud import Client

client = Client.from_config()
solver = client.get_solver(supported_problem_types__contains='qcdl')

qcdl = {...}
params = dict(shots=3, time_limit=600, noisy=False)

response = solver.sample_qcdl(qcdl, **params)
answer = orjson.loads(response.answer_data.read())

Alternatively, to upload the circuit first, replace the sample_qcdl line with:

circuit_id = solver.upload_qcdl(qcdl).result()
response = solver.sample_qcdl(circuit_id, **params)

AI Generation Disclosure

No AI used.

[codecov]

Codecov Report

❌ Patch coverage is 92.85714% with 1 line in your changes missing coverage. Please review.
✅ Project coverage is 89.93%. Comparing base (2261ca5) to head (d831938).

Files with missing lines	Patch %	Lines
dwave/cloud/solver.py	90.00%	1 Missing ⚠️

Additional details and impacted files

@@           Coverage Diff           @@
##           master     #745   +/-   ##
=======================================
  Coverage   89.92%   89.93%           
=======================================
  Files          44       44           
  Lines        5300     5313   +13     
=======================================
+ Hits         4766     4778   +12     
- Misses        534      535    +1     

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[arcondello]

I have always loved this design. No more and no less than the amount of information needed to support the new problem type.

[arcondello]

I wonder if a try/except here might help some users debug in the future.

OTOH we don't have evidence that this is a common failure path, so until we do probably better to leave it alone. Ok, I talked myself out of it, but leaving the comment for posterity 😄

[randomir]

Once we integrate QCDL generation into Ocean, we'll be able to do better checks here. Or even defer serialization completely. Just like we do with NL (and require dwave-optimization) or BQM/CQM/DQM (and require dimod).