Architecture design for Spark v2 running allongside Spark v1

[bajtos]

As explained in https://www.notion.so/spacemeridian/Spark-v2-design-migration-115cdd5cccdb804ca4d1e0694c613318, we need to run Spark v2 alongside Spark v1 to give the ecosystem enough time to adapt to the new requirements and create space for us to fix any unforeseen problems in the way how Spark v2 integrates with Filecoin and IPNI.

To do so, we need an architecture design doc and a high-level implementation plan how to get from the current state, where we have only one subnet running (Spark v1), to the new state, where we have two versions of Spark running alongside each other.

The section Overview of components documents which Spark components can share the same code and which must be adapted for Spark v2.

Expected outcomes:

• architecture proposal/design doc
• high-level implementation plan