Deprecate mapParallel procedure and update docs

[conker84]

trafficstars

Deprecate the procs and update the docs by explaining how to move to the Parallel runtime

[alexanoid]

Please do not delete apoc.cypher.mapParallel2 because at the moment, the new PARALLEL runtime cannot fully replace it. Details https://stackoverflow.com/questions/77509091/neo4j-parallel-runtime-query-returns-different-result

Example: I have many scenarios where I currently use apoc.cypher.mapParallel2 to read data in the same transaction where I'm updating it. I would like to transition to the PARALLEL runtime, but I am unable to do so due to this limitation.

Also, I'm testing the new Neo4j PARALLEL runtime and am unable to pass my tests because my queries return different results (different elements order) compared to SLOTTED or PIPELINED.

[alexanoid]

For example I'm unable to port my query which is based right now on apoc.cypher.mapParallel2 to PARALLEL runtime https://stackoverflow.com/questions/77814894/neo4j-on-parallel-runtime-returns-different-results-different-elements-order-c

[alexanoid]

Additionally, with the help of apoc.cypher.mapParallel2, I can parallelize only a specific part of the query. This is a feature which is not available with the PARALLEL runtime.

With apoc.cypher.mapParallel2 I may use chaining with parallel and non-parallel executions in one single query. Please keep this procedure in the library and leave us a choice where to use PARALLEL runtime and where apoc.cypher.mapParallel2.