chore: remove committed local artifacts and tidy .gitignore

[egeboy35]

This PR removes a few files from version control that appear to be locally generated rather than authored, and tidies up .gitignore so they stay ignored going forward.

Problem

A handful of machine-generated files are currently tracked: a root-level .DS_Store (already listed in .gitignore, but tracked from before the rule was added), an empty tx_log.txt, a ~91 KB upload_log.txt (an HTTP transfer log written at runtime by rf_stream/upload_to_hf.py), and the MATLAB/Simulink JIT/codegen cache under sdrmatlab/slprj/. These change on every local run and add noise to diffs. Separately, .gitignore had accumulated a few duplicate entries.

Fix

• Remove the files above from version control. sdradi/pluto_test/test_video.mp4 is untracked from the index only (git rm --cached), since several sdradi scripts use test_video.mp4 as their default input; it was already covered by an ignore rule added in the same commit as the file. Note that pulling this change removes the working-tree copy of that file (standard git behavior when untracking), so you may want to keep a local copy if you still use it.
• .gitignore: drop duplicate entries (__pycache__/, *.egg-info/, build/, dist/, .env), consolidate the three rf_stream_rx_runs variants into the single unanchored entry, and remove the stock-template bare lib/ entry. Add slprj/, tx_log.txt, upload_log.txt, *.pth, and *.ckpt. The intentional docs and newsdr entries are left as they are.
• imgs/64QAM_copy.png was deliberately left alone: it differs from imgs/64QAM.png and is written by sdrpysim/testmatplotlibcv2.py.

Verification

git check-ignore confirms every removed path is matched by the resulting .gitignore; git status is clean after the commit and git ls-files no longer lists the removed paths. Happy to adjust or split this if you'd prefer any of these handled differently.

🤖 Generated with Claude Code

[egeboy35]

Correcting one detail in my own description above: the rationale I gave for using git rm --cached on sdradi/pluto_test/test_video.mp4 names the wrong file, and the real situation makes the removal safer rather than riskier.

I wrote that the file is kept on disk "since several sdradi scripts use test_video.mp4 as their default input". Those scripts do exist — ten of them, e.g. run_video_txv2.py:111, run_loopback.py:141, sim_video_e2e_asyncv3.py:1284 — but every reference is the bare relative name "test_video.mp4", so it resolves against the working directory, i.e. sdradi/test_video.mp4 when run from sdradi/. That path is a different file, and it is itself ignored (.gitignore:181) and untracked, so it only exists as a local copy. Nothing under sdradi/pluto_test/ references the tracked copy either.

So the tracked pluto_test copy is not the file those scripts load, and untracking it does not affect them. The only real consequence stands as stated: pulling this change removes that file from your working tree, so if you use it directly from pluto_test/, keep a local copy first.

🤖 Generated with Claude Code