Slim fixture: mic-smolvlm-slim

A 6.6M-parameter fixture for the smolvlm MIC, covering HuggingFaceTB/SmolVLM2-2.2B-Instruct. It is a test artifact, not a usable model, tagged test-artifact to keep it out of candidate-model search.

• Fixture key 9530c0e74c36. It covers every checkpoint whose processor files are byte-identical and whose config matches after shrinking.
• Same architecture, smaller config: SmolVLMForConditionalGeneration, shrunk by the generic rule in sweep.py (vision num_hidden_layers 2, text num_hidden_layers 2).
• Processor files copied byte for byte from HuggingFaceTB/SmolVLM2-2.2B-Instruct: added_tokens.json, chat_template.json, generation_config.json, merges.txt, preprocessor_config.json, processor_config.json, special_tokens_map.json, tokenizer.json, tokenizer_config.json, vocab.json.
• Random weights (seed 17). They passed the input-sensitivity gate: every perturbation of the contract's inputs moved the logits by at least 0.001 (weakest: attention_mask:all_ones, 0.00274), so no training was needed.
• Same tensor names as the real checkpoint. The real checkpoint has every parameter, and so does the fixture.
• Saved in float32, the dtype HuggingFaceTB/SmolVLM2-2.2B-Instruct loads in by default.
• Finite in float32, bfloat16, and float16 (forward, loss, gradients).

Built by build_fixtures.py; build.json records the pinned sources, gate results, and file hashes. The source repository commits only README.md and build.json; rebuild the other files with python example/build_fixtures.py --fixture mic-smolvlm-slim.