[MEL] Use message extractor to completely replace inbox reader and tracker code

[ganeshvanahalli]

This PR aims to get rid of pending inbox reader and tracker dependency of nitro nodes, and instead to start using message extractor.

Resolves NIT-4753

[codecov]

Codecov Report

❌ Patch coverage is 48.94737% with 97 lines in your changes missing coverage. Please review.
✅ Project coverage is 34.30%. Comparing base (40eec39) to head (9d955a1).
⚠️ Report is 30 commits behind head on master.

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #4593      +/-   ##
==========================================
+ Coverage   34.12%   34.30%   +0.18%     
==========================================
  Files         497      497              
  Lines       59178    59233      +55     
==========================================
+ Hits        20193    20319     +126     
+ Misses      35432    35373      -59     
+ Partials     3553     3541      -12     

[github-actions]

❌ 5 Tests Failed:

Tests completed	Failed	Passed	Skipped
4801	5	4796	0

View the top 3 failed tests by shortest run time

TestDataStreaming_PositiveScenario/Many_senders,_long_messages

Stack Traces | 0.160s run time

=== RUN   TestDataStreaming_PositiveScenario/Many_senders,_long_messages
WARN [04-10|05:16:22.263] Served datastreaming_start               conn=127.0.0.1:40846 reqid=6 duration="111.84µs" err="we have already seen this request; aborting replayed protocol"
INFO [04-10|05:16:22.273] rpc response                             method=datastreaming_start logId=6 err="we have already seen this request; aborting replayed protocol" result={} attempt=0 args="[\"0x69d887a6\", \"0x31\", \"0xd9\", \"0x28bb\", \"0xa\", \"0x900690047222078d1be3f6874984e4bfbc9e908b10edb6019d9d60fd7e00d841702521c77a9ecfe43d5c49011eb790c6297621db9204c83b29da75b15f49f69f00\"]" errorData=null
    protocol_test.go:230: goroutine 282 [running]:
        runtime/debug.Stack()
        	/opt/hostedtoolcache/go/1.25.8/x64/src/runtime/debug/stack.go:26 +0x5e
        github.com/offchainlabs/nitro/util/testhelpers.RequireImpl({0x164b490, 0xc000183500}, {0x16319c0, 0xc000802960}, {0x0, 0x0, 0x0})
        	/home/runner/work/nitro/nitro/util/testhelpers/testhelpers.go:29 +0x9f
        github.com/offchainlabs/nitro/daprovider/data_streaming.testBasic.func1()
        	/home/runner/work/nitro/nitro/daprovider/data_streaming/protocol_test.go:230 +0x19b
        created by github.com/offchainlabs/nitro/daprovider/data_streaming.testBasic in goroutine 274
        	/home/runner/work/nitro/nitro/daprovider/data_streaming/protocol_test.go:223 +0x85
        
    protocol_test.go:230: �[31;1m [] we have already seen this request; aborting replayed protocol �[0;0m
--- FAIL: TestDataStreaming_PositiveScenario/Many_senders,_long_messages (0.16s)

TestDataStreaming_PositiveScenario

Stack Traces | 0.210s run time

=== RUN   TestDataStreaming_PositiveScenario
--- FAIL: TestDataStreaming_PositiveScenario (0.21s)

TestAnyTrustRekey

Stack Traces | 2.410s run time

... [CONTENT TRUNCATED: Keeping last 20 lines]
TRACE[04-10|05:18:59.988] Engine API request received              method=GetPayload                    id=0x032966b4657db2b5
DEBUG[04-10|05:18:59.987] Served eth_unsubscribe                   reqid=1671 duration="11.481µs"
DEBUG[04-10|05:18:59.988] Pushed sync data from consensus to execution synced=true  maxMessageCount=14  updatedAt=2026-04-10T05:18:59+0000 hasProgressMap=false
DEBUG[04-10|05:18:59.988] Pushed sync data from consensus to execution synced=true  maxMessageCount=3   updatedAt=2026-04-10T05:18:59+0000 hasProgressMap=false
DEBUG[04-10|05:18:59.988] Served eth_getBlockByNumber              reqid=198  duration="84.046µs"
DEBUG[04-10|05:18:59.988] Served eth_getTransactionCount           reqid=1561 duration="49.523µs"
DEBUG[04-10|05:18:59.988] Served eth_getBlockByNumber              reqid=1562 duration="24.656µs"
DEBUG[04-10|05:18:59.988] Pushed finality data from consensus to execution safeMsgCount=14 finalizedMsgCount=1 validatedMsgCount=0
TRACE[04-10|05:18:59.988] Handled RPC response                     reqid=1561 duration="1.052µs"
DEBUG[04-10|05:18:59.988] Executing EVM call finished              runtime="200.023µs"
DEBUG[04-10|05:18:59.988] Served eth_maxPriorityFeePerGas          reqid=173  duration="168.613µs"
DEBUG[04-10|05:18:59.988] Served eth_call                          reqid=389  duration="230.25µs"
TRACE[04-10|05:18:59.988] Handled RPC response                     reqid=1562 duration="1.113µs"
TRACE[04-10|05:18:59.988] Handled RPC response                     reqid=173  duration="1.072µs"
DEBUG[04-10|05:18:59.987] Served eth_blockNumber                   reqid=693  duration="22.392µs"
TRACE[04-10|05:18:59.988] Handled RPC response                     reqid=389  duration="1.383µs"
TRACE[04-10|05:18:59.988] Handled RPC response                     reqid=198  duration="1.253µs"
DEBUG[04-10|05:18:59.987] Served eth_unsubscribe                   reqid=694  duration="15.218µs"
--- FAIL: TestAnyTrustRekey (2.41s)
TRACE[04-10|05:18:59.988] Handled RPC response                     reqid=1638 duration="2.284µs"

📣 Thoughts on this report? Let Codecov know! | Powered by Codecov

[rauljordan]

Hi @ganeshvanahalli I pushed some fixes from PR review toolkit which found the following real issues:
Code Fixes (5 files)
Critical #1
Added nil guard for InboxReader in GetParentChainDataSource() to prevent panic when both MessageExtractor and InboxReader are nil.

Critical #3
Changed return 0, nil to return m.config.RetryInterval, nil after preimage cache rebuild. Prevents unbounded CPU spin when the preimage is permanently missing from DB.

High #4
Replaced blocking channel send m.reorgEventsNotifier <- ... with a select on ctx.Done() to prevent deadlocking the MEL FSM if the reader goroutine is slow/stopped.

High #5
Extracted the error into a package-level sentinel errFindDelayedNotImplementedByMEL and downgraded the caller's log from Warn to Debug (also logging the error itself). Eliminates log spam on every message retrieval when MEL is active.

High #6
Moved ChainId validation before the BlobSchedule == nil early return, so chain ID mismatches are always detected even when BlobSchedule is nil.

Unit Tests Added 27 new test cases in mel_test.go

[eljobe]

This is all Claude-generated PR review feedback.
I did cull it down from a longer list that didn't look like as much high-quality feedback.

Please take a look and see what can be addressed in this PR.