Enable stable Android concurrent BTC payments e2e test

[MaiborodaY]

Summary

Re-enables org.rgblightningnode.ConcurrentBtcPaymentsTest in the Android E2E CI matrix.

Validation

• Ran ConcurrentBtcPaymentsTest locally on Android emulator Pixel_API_34.
• Repeated 10 runs with clean regtest restart before each run.
• Result: 10/10 passed.

Notes

Other disabled Android E2E tests remain disabled:

• RestartTest: tracked separately as a restart/unlock persistence issue #771 https://github.com/UTEXO-Protocol/project-tasks/issues/771
• MultiOpenCloseTest and SwapRoundtripBuyTest: blocked by RGB openchannel race causes funding flow panic before RGB metadata is persisted #28

[MaiborodaY]

Update on `org.rgblightningnode.ConcurrentBtcPaymentsTest`:

I re-enabled it after local validation, but CI still failed on the Android E2E job.

Local result:
- 40 repeated runs on `Pixel_API_34`
- clean regtest restart before each run
- 40/40 passed

CI result:
- failed while waiting for the first BTC channel to become usable
- the funding tx was found and confirmed
- failure happened in `waitForUsableChannel(...)`

Visible CI error:

InternalException: called Result::unwrap() on an Err value: PoisonError { .. }
at SdkNode.sync(...)
at ConcurrentBtcPaymentsTest.waitForUsableChannel(...)
at ConcurrentBtcPaymentsTest.concurrentBtcPayments(...)

Relevant CI flow:

node B connectpeer: ok
node C connectpeer: ok
node D connectpeer: ok
channel funding tx found: 56ee2c0e...
funding tx confirmed in block: 56ee2c0e...
mined 6 block(s)
waiting for usable channel... (poll 1)
PoisonError at SdkNode.sync(...)

This is different from the RGB openchannel/coloring issue seen in `MultiOpenCloseTest` / `SwapRoundtripBuyTest`, because this failure happens on a BTC-only channel open (`assetId = null`, `assetAmount = null`).

I’ll keep investigating this in the same branch. For now, `ConcurrentBtcPaymentsTest` should not be re-enabled until we find the primary Rust-side panic behind the `PoisonError`.