Scale typo gates with length and refuse destructive auto-runs

Similarity requirements now scale with word length (typo.floor_ok:
single edit or ratio >= max(0.6, 1 - 3/max(len,3)), first char kept;
zsh spdist / nushell did_you_mean derivation in the source). The
fixed-cutoff gates are swapped: history_resolver._similar ->
floor_ok with the divergence cap deleted (every diverged token is
gated per-token, not counted), learned.guess_from_path and the help
resolver candidate gates -> floor_ok, so _TOKEN_CUTOFF,
_MAX_DIVERGED, GUESS_CUTOFF, _CUTOFF and their difflib plumbing are
gone; unique-survivor, no-op and which/'/'/'.'/flag guards kept.

New thefuck/danger.py is_dangerous(script) parses via bashlex
directly and fail-safes to True when bashlex is unavailable or the
script refuses to parse (the flat fallback is head-only); with a
tree it matches rm/rmdir recursive flags, dd/mkfs*/shred/wipefs/
mkswap heads, git push --force/-f (not --force-with-lease),
chmod/chown -R with a 777-style mode, kill -9, fork-bomb shapes,
pipe-to-shell tails and file redirects outside /tmp and /dev/null.
fix_command checks it before ANY auto-run, learned-db exact hits
included: dangerous candidates fall through to rules+ask.

Test-migration inventory (authorized semantic inversions):
- tests/resolvers/test_history_resolver.py: declines-3-diverged ->
  corrects (cap deleted); 0.8-cutoff boundary arithmetic re-based to
  floor boundaries (len-3 0.6 / len-10 0.7 / len-30 0.9);
  _TOKEN_CUTOFF import removed with the constant; declines-just-
  below-cutoff re-based to the len-20 floor 0.85; added a 17-char
  below-floor decline.
- tests/test_learned.py: returns-none-below-cutoff re-based to a
  same-first-char below-floor pair (0.6 < len-10 floor 0.7); added a
  ratio-0.7 acceptance pin; single-edit-under-cutoff renamed.
- tests/resolvers/test_help_resolver.py: transposition comments
  re-based to the floor; added a below-floor subcommand decline.
- tests/entrypoints/test_fix_command_learned.py: exact-learned-wins
  fixture's 'git push --force' correction (now correctly refused)
  replaced by benign scripts; mock_learned stubs danger benign for
  platform-neutral auto-apply tests; added TestDangerOverride
  (real module, all four sources: reaches select_command, nothing
  auto-runs).

tests/resolvers/test_help_resolver.py

@@ -96,21 +96,31 @@ def test_returns_correction_when_subcommand_misspelled(
 
 
 def test_corrects_transposed_subcommand(fake_bin, argv_log, enable_cache):
-    # bulid -> build scores 2 * 4 / 10 = 0.8, right at the cutoff,
-    # and is also a single transposition (li <-> il).
+    # bulid -> build is a single adjacent transposition (li <-> il),
+    # so the length-scaled floor admits it on the single-edit path.
     fake_bin('fake')
     assert get_help_correction('fake bulid .') == 'fake build .'
     assert typo.single_edit('bulid', 'build')
 
 
-def test_corrects_transposition_under_cutoff(
+def test_corrects_transposition_via_single_edit(
         fake_bin, argv_log, enable_cache):
-    # psuh -> push scores 2 * 3 / 8 = 0.75, under the cutoff, so only
-    # the first-char-equal single-edit match fixes it.
+    # psuh -> push scores 2 * 3 / 8 = 0.75, so only the single-edit
+    # path of the floor admits it.
     fake_bin('fake', commands=('push', 'build', 'clean'))
     assert get_help_correction('fake psuh x') == 'fake push x'
 
 
+def test_declines_subcommand_below_scaled_floor(
+        fake_bin, argv_log, enable_cache):
+    # 10-char subcommand vs the listed 'burnno1234' scores
+    # 2 * 6 / 20 = 0.6, under the len-10 floor 0.7, and is not a
+    # single edit; no listed subcommand is close enough.
+    fake_bin('fake', commands=('burnno1234', 'build', 'clean'))
+    assert get_help_correction('fake burnnoabcd') is None
+    assert _spawned(argv_log) == ['--help']
+
+
 def test_respawns_when_binary_mtime_changes(
         fake_bin, argv_log, enable_cache):
     path = fake_bin('fake')