cc_seed_random: update documentation and fix integration test

[OddBloke]

Proposed Commit Message

cc_seed_random: update documentation and fix integration test

The documentation did not mention that the given data may not be the
exact string written: the cloud's random data may be added to it.
Additionally, the documentation of the command key was incorrect.

test_seed_random_data was updated to check that the given data is a
prefix of the written data, to match cloud-init's expected (and, now,
documented) behaviour.

LP: #1911227

Test Steps

Run CLOUD_INIT_PLATFORM=azure pytest tests/integration_tests/modules/test_seed_random_data.py and observe a pass (vs. a fail previously).

Checklist:

• My code follows the process laid out in the documentation
• I have updated or added any unit tests accordingly
• I have updated or added any documentation accordingly

[smoser]

I'm good with this as it is or with any amount of my feedback taken. (But I do think the paren are not necessary 😉).

[smoser]

(the parens are not necessary here). It also wasn't completely clear to me what or where "data was written". That is possibly because I was just reading this as diff output, so maybe I lost context.

Maybe:

If a ``command`` key is present, the command will be executed after ``file`` has been populated.
The command's environment will contain the value of ``file`` specified above in the key 
``RANDOM_SEED_FILE``. If a command is specified that cannot be run, no error will 
be reported unless ``command_required`` is set to true.