Disabled two tests in recording_micro_allocator_test.cc

[veblush]

This is a temporary workaround to resolve a test failure on the main branch. This change should be reverted once the underlying issues are fixed and the tests are re-enabled.

The following tests are failing when executed with qemu-arm:

• TestRecordsPersistentTfLiteTensorData
• TestRecordsPersistentTfLiteTensorQuantizationData

Repro: (From the TFLM root directory)

$ docker run -it --rm -v $(pwd):/opt/tflm -w /opt/tflm ghcr.io/tflm-bot/tflm-ci:0.6.5  /bin/bash
# tensorflow/lite/micro/tools/ci_build/test_cortex_m_qemu.sh
# qemu-arm -cpu cortex-m3 gen/cortex_m_qemu_cortex-m3_default_cmsis_nn_gcc/bin/recording_micro_allocator_test

BUG=QuickWorkaroundOnMain

[ddavis-2015]

@veblush I am suspicious of the existing kTestConvModelData (a generated convolutional model). The model we are using is from 2021, and when regenerated using the repo script, generates a very different model today. In my PR for schema buffer and custom option offsets, I generated a new model and placed it into the PR.

[veblush]

Thanks for the pointer. I also concluded that the pre-generated test model file is misaligned, which is the culprit. I'm regenerating it in a separate PR to confirm this.

[veblush]

See #3207

[mergify]

This pull request has been removed from the queue for the following reason: checks timeout.

The merge conditions cannot be satisfied due to checks timeout.

You can check the last failing draft PR here: #3208.

You may have to fix your CI before adding the pull request to the queue again.

If you update this pull request, to fix the CI, it will automatically be requeued once the queue conditions match again.
If you think this was a flaky issue instead, you can requeue the pull request, without updating it, by posting a @mergifyio requeue comment.