Standardize model filenames and delete legacy model

## Changes

### Renames (standardized naming: {model}-genie-{quant}-{chip}.tar.gz)
- llama-v3.2-1b-instruct-* → llama3.2-1b-instruct-* (drop v prefix)
- llama-sea-lion-v3.5-8b-r-* → sea-lion3.5-8b-instruct-* (clean name)
- qwen2.5-7b-instruct-genie-w8a16 → qwen2.5-7b-instruct-genie-w8a16-8elite (add chip suffix)

### Deleted
- llama-3.2-1b-instruct-genie-w4.tar.gz (legacy, replaced by w4a16 variants)

### Unchanged
- qwen3-4b-genie-w4a16-8elite-gen5.tar.gz (already correct)

llama-3.2-1b-instruct-genie-w4.tar.gz

@@ -1,3 +0,0 @@
-version https://git-lfs.github.com/spec/v1
-oid sha256:0f936d1b54765eebc159dd8de18431fa8c77a7f4bf45e0199235499950a1abd4
-size 1081009536