fix: kill old execution module when redeploying under same namespace and identifier

[deekerno]

Changelog

• Kill old execution module if deploying a new execution module under the same namespace and identifier

Reproduction (on master)

0. Ensure database is clear and build forc-index: cargo build -p forc-index.
1. Add log message to examples/block-explorer/explorer-index/src/lib.rs; something like "FIRST" will suffice.
2. Deploy explorer index:

../../target/debug/forc-index deploy \                                                      
   --path explorer-index \
   --output-dir-root [whatever your fuel-indexer repo root is] \
   --url http://0.0.0.0:29987 \
   --target wasm32-unknown-unknown

3. Change log message in explorer example to something different, e.g. "SECOND".
4. Re-deploy using same command.
5. Search through service logs; you should see that both messages are being logged.

Testing Plan (on this branch)

Repeat the above steps, you should see that only the second message is being logged. There should also be a message saying the following:

Replaced execution module for fuel_examples.explorer_index, stopping old execution module.

[ra0x3]

• utACK
• @deekerno nice find here 🙏🏽
• Left a comment. We should definitely refer to this as an executor (the literal execution coroutine).
• We should also refer to the manifest.uid() as an "indexer"
  • There are some other parts of the code we need to update with this as well but this can be done separately. Basically need to replace all references to "indexer" as "indexer" (unless it's referring to the literal data that is indexed by an indexer - which is indeed an index)