Clarify Function Names for Logging

[Satrat]

SUMMARY:

• Each module compressor(sparseGPT, WANDA, GPTQ) implemented a fasterprune method, but this is misleading for GPTQ which does quantization only. Renaming fasterprune->compress
• Rename LayerCompressor.prune() to LayerCompressor.compress_module() for the same reason

TEST PLAN:
Manually tested a w4a16 quantization example to confirm the logging change:

Before

2024-07-02T14:15:00.233030+0000 | prune | INFO - Compressing model.layers.11.self_attn.q_proj.model.layers.11.self_attn.q_proj...
2024-07-02T14:15:01.061103+0000 | fasterprune | INFO - time 0.83
2024-07-02T14:15:01.061573+0000 | fasterprune | INFO - error 6155.81

After

2024-07-02T14:31:16.224969+0000 | compress_module | INFO - Compressing model.layers.20.model.layers.20.self_attn.q_proj...
2024-07-02T14:31:17.100250+0000 | compress | INFO - time 0.87
2024-07-02T14:31:17.100596+0000 | compress | INFO - error 4623.34