Issue 133: correct the stale RolledOutIntervals skip reason

[bryantaustin13]

Addresses #133.

What

Replaces the stale RolledOutIntervals skip reason in the four shipped configs that carried it, and
fixes a JSON syntax error in one of them.

The old reason cited an identifier the test has not contained since 2024-08:

CQLtoELM - Could not resolve identifier MedicationRequestIntervals in the current library

The new reason is the error the test actually produces today, verified against the reference engine
(HAPI FHIR 8.10.0 / CQF engine 4.1.0, $cql, CQL 1.5) using the expression exactly as the runner
sends it:

CQLtoELM - Expected an expression of type 'decimal', but found an expression of type
'System.Integer' in Quantity{ value: duration in days of X }. Tracked in
https://github.com/cqframework/cql-tests-runner/issues/133

Full engine diagnostic:

Expected an expression of type 'decimal', but found an expression of type 'System.Integer'.,
Could not determine signature for invocation of operator System.Add.,
Internal translator error., Internal translator error.,
Could not determine signature for invocation of operator System.Union.

Files: conf/development.json, conf/localhost-evaluate.json, conf/localhost.json,
conf/smile-cdr-local.json.

conf/smile-cdr-local.json additionally had a trailing comma after "cqlEngineVersion", which made
it invalid JSON — ConfigLoader threw on it, so that configuration could not be loaded at all. Fixed
here because the file could not otherwise be edited or verified.

Why it matters

These configurations ship with the runner, so the obsolete message is reproduced in every recorded
run in cql-tests-results — cqf-java, cqf-javascript, firely and cozeva all report:

"testStatus": "skip",
"skipMessage": "Skipped by config: CQLtoELM - Could not resolve identifier MedicationRequestIntervals in the current library"

Anyone investigating that skip chases an identifier that is not in the test.

Relationship to cqframework/cql-tests#146

This PR is not blocked by #146. #133 proposes eventually deleting these skip entries, and that
does depend on #146 ("Fix RolledOutIntervals starting type") merging plus a submodule bump. This PR
does not delete them — it corrects a reason that misdescribes reality, which is independent of #146
and can merge now.

Worth noting for the follow-up: the test still fails to translate both with the current
List<Interval<DateTime>> starting type and with #146's List<Interval<Date>>, because the
Decimal/Integer error above is unrelated to the starting type. So #146 is necessary but not
sufficient for removing these skips — the entries cannot simply be deleted once it merges.

Not included

conf/cql-execution-local.json also skips this test, but for a current and correctly attributed
reason linking #80. That entry is left alone; it should be re-evaluated separately now that #129 has
merged List<...> unwrapping for cqf-cqlType, and that needs an engine this change was not
verified against.

Verification

• All five configurations parse as JSON and load through ConfigLoader (previously
  smile-cdr-local.json did neither).

• Full suite against the reference engine: 1639 pass / 170 fail / 14 skip / 0 error, unchanged
  from main. The test remains skipped, now reporting the corrected reason:

  Test CqlAggregateTest:AggregateTests:RolledOutIntervals status: skip
    skipMessage: Skipped by config: CQLtoELM - Expected an expression of type 'decimal', ...
  

• No occurrence of MedicationRequestIntervals remains in conf/ or src/.

• Run on Node 26 (engines: >=26) with a clean dependency install; results identical to the earlier
  Node 22 run.