[rl][ci] Isolate CI hang: disable flex bitwise-parity test + fix stale RL H100 badge

- Comment out TestBitwiseParityFlex torchrun step: the flex generator hits the
  vLLM flex_attention_compiled recompile-limit deadlock (#3898), which hangs the
  A10G RL integration job until CI times out. Disabled to confirm flex is the
  hang source (varlen teardown was fixed separately).
- Remove the dead RL H100 CI badge from experiments/README.md (the workflow was
  removed in #3855); it 404s and fails the lychee link checker.

Author: wwwjn

.github/workflows/integration_test_8gpu_rl.yaml

@@ -121,9 +121,12 @@ jobs:
         RL_TEST_DUMP_FOLDER="$RUNNER_TEMP/artifacts-to-be-uploaded" \
           HF_ASSETS_PATH="$MODEL_PATH" torchrun --nproc-per-node=2 -m pytest \
           torchtitan/experiments/rl/tests/test_bitwise_parity.py::TestBitwiseParityVarlen -v
-        RL_TEST_DUMP_FOLDER="$RUNNER_TEMP/artifacts-to-be-uploaded" \
-          HF_ASSETS_PATH="$MODEL_PATH" torchrun --nproc-per-node=2 -m pytest \
-          torchtitan/experiments/rl/tests/test_bitwise_parity.py::TestBitwiseParityFlex -v
+        # TODO(#3898): re-enable once the flex generator recompile-limit hang is
+        # fixed (vLLM flex_attention_compiled recompiles per query length and
+        # deadlocks on A10G). Temporarily disabled to isolate the CI hang.
+        # RL_TEST_DUMP_FOLDER="$RUNNER_TEMP/artifacts-to-be-uploaded" \
+        #   HF_ASSETS_PATH="$MODEL_PATH" torchrun --nproc-per-node=2 -m pytest \
+        #   torchtitan/experiments/rl/tests/test_bitwise_parity.py::TestBitwiseParityFlex -v
 
         # Run E2E RL integration tests (up to 8 GPUs): includes the MoE TP=4 EP=4
         # random-init test and the batch-invariant debug tests (both need 8 GPUs).

torchtitan/experiments/README.md

@@ -27,5 +27,5 @@ We provide this `experiments/` folder to host experiments that add significant v
 | [graph_trainer](./graph_trainer/) | [![GraphTrainer 8 GPU Integration Tests](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_graph_trainer.yaml/badge.svg?branch=main)](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_graph_trainer.yaml?query=branch%3Amain) [![GraphTrainer H100 8 GPU Integration Tests](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_graph_trainer_h100.yaml/badge.svg?branch=main)](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_graph_trainer_h100.yaml?query=branch%3Amain) | [@SherlockNoMad](https://github.com/SherlockNoMad) |
 | [torchft](./torchft/) | [![TorchFT 8 GPU Integration Tests](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_torchft.yaml/badge.svg?branch=main)](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_torchft.yaml?query=branch%3Amain) | [@tushar00jain](https://github.com/tushar00jain) [@fegin](https://github.com/fegin) |
 | [transformers_modeling_backend](./transformers_modeling_backend/) | [![Transformers modeling backend 8 GPU Integration Tests](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_transformers_modeling_backend.yaml/badge.svg?branch=main)](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_transformers_modeling_backend.yaml?query=branch%3Amain) | [@3outeille](https://github.com/3outeille) [@mreso](https://github.com/mreso) |
-| [rl](./rl/) | [![RL 8 GPU Integration Tests](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_rl.yaml/badge.svg?branch=main)](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_rl.yaml?query=branch%3Amain) [![RL H100 8 GPU Integration Tests](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_rl_h100.yaml/badge.svg?branch=main)](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_rl_h100.yaml?query=branch%3Amain) | [@wwwjn](https://github.com/wwwjn) [@felipemello1](https://github.com/felipemello1) |
+| [rl](./rl/) | [![RL 8 GPU Integration Tests](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_rl.yaml/badge.svg?branch=main)](https://github.com/pytorch/torchtitan/actions/workflows/integration_test_8gpu_rl.yaml?query=branch%3Amain) | [@wwwjn](https://github.com/wwwjn) [@felipemello1](https://github.com/felipemello1) |
 | [forge](./forge/) | TBA | [@felipemello1](https://github.com/felipemello1) |