Invalid Span Extracted when using XrayPropagator

Describe your environment
Python 3.11
FastAPI 0.103.0
opentelemetry-instrumentation-fastapi==0.41b0
opentelemetry-propagator-aws-xray==1.0.1
opentelemetry-sdk==1.20.0

Steps to reproduce

Minimal example.

propagate.set_global_textmap(AwsXRayPropagator())
resource = Resource.create(attributes={"service.name": "test-service"})
tracer = TracerProvider(resource=resource)

jaeger_exporter = JaegerExporter(
    agent_host_name="localhost",
    agent_port=6831,
)
processor = BatchSpanProcessor(jaeger_exporter)
tracer.add_span_processor(processor)
samp = ParentBased(ALWAYS_ON)

trace.set_tracer_provider(
    TracerProvider(
        resource=resource,
        active_span_processor=processor,
        id_generator=AwsXRayIdGenerator(),
        sampler=samp
    )
)

tracer = trace.get_tracer(__name__)

app = FastAPI(title="test api", version="draft")
FastAPIInstrumentor().instrument_app(app, tracer_provider=trace.get_tracer_provider())

@app.get("/system/ping")
async def ping() -> dict:
    return {"message": "pong"}


if __name__ == '__main__':
    uvicorn.run(app, host='0.0.0.0', port=8081,
        log_config='logging_conf.yml',
        workers=1)

What is the expected behavior?
Passing in the below header as an example.
X-Amzn-Trace-Id: Root=1-5759e988-bd862e3fe1be46a994272793

I would expect Xray to join up traces across services as a linked trace or part of the same trace

What is the actual behavior?
Traces coming through to Xray unlinked and the following error.

Invalid Span Extracted. Inserting INVALID span into provided context.

Start by reproducing the issue with the supplied FastAPI example and X-Amzn-Trace-Id header, then trace XrayPropagator's extraction path where the invalid span is reported. Done means incoming X-Ray context is accepted so spans are linked across services and the invalid-span error no longer occurs.