Add support for multiple processor instances with different configuration options within a workflow

[nevali]

e.g.,

workflow=sparql-get:source,sparql-put,s3-put:mybucket

[s3-put:mybucket]
;; specific options for this processor being invoked with the "mybucket" instance name

Specifying a name which lacks a configuration section would be an error.

This could be achieved through a simple API in libtwine that modules call during initialisation to scan the config for instances of themselves and register them as processors

Tracked as RESDATA-1289