add support for requesting parallel engines in a http batch job

[giorgosHadji]

To specify the number of ASR engines a job on http batch worker(on Prem container) can use we added a new header called X-SM-Processing-Data. Inside the header we can have a json dictionary, with a key called requested_parallel where you can specify the number of engines you want to use.

Usage with:

`from speechmatics.batch import AsyncClient

async def main():
async with AsyncClient(url='http://localhost:9002/v2') as client:
result = await client.transcribe(sys.argv[1],requested_parallel=2)
print(result)

asyncio.run(main())`