feat(#major); abracadabra; subgraph optimizations

[dhruv-chauhan]

• Huge arrays in abracadabra-money-avalanche deployment causing issues for other subgraphs in the shard.
• As a suggested storage optimization, we looked to convert large arrays (length>50) as @derivedFrom mappings, but on querying it looks like the larger arrays already follow this convention, others do not need to due to very small length. Reference query; schema.
• Since most of these large arrays are snapshots and events, we could further make these entities immutable.
• Changed ID fields from strings to Bytes.
• Test deployment: https://okgraph.xyz/?q=dhruv-chauhan%2Fabracadabra-money-avalanche

[melotik]

code changes lgtm and the schema looks good. Just want to confirm the comment I made. If the IDs change at all we have assumptions downstream that need to be coordinated