FIX float counts in Paladin mapping (#74)

* FIX setup.py

- Eliminate problems of compability

* Add AMP and Hemo models for compatibility

- Add downgraded IR versions of Hemo and AMP models to wide compatibility
- Downgrade of IR versions to 7 (allows compatibility to the present the installed environment and before ones)

Include the same models adapted with:

```
>>> from onnx import version_converter
>>> model = onnx.load('TEST.onnx')
>>> model.ir_version = 7
>>> model = optimize(model, ["eliminate_identity", "fuse_consecutive_transposes"])
>>> model = version_converter.convert_version(model, 16)
>>> onnx.save(model, "TEST.onnx")
```

* ENH count.ngl

Include a filter of sam matchers:

filtered = select(input, keep_if=[{unique}])

and a count parameter to avoid secondary aligments:

    multiple={unique_only}

Author: celiosantosjr

macrel/data/models/AMP.onnx.gz

@@ -3,13 +3,14 @@ ngless "1.0"
 input = samfile(ARGV[1],
              name="MACREL")
 
+# Keep only primary alignments
 filtered = select(input, keep_if=[{unique}])
 
+# Count features from filtered alignments
 counts = count(filtered,
     features=["seqname"],
     include_minus1=False,
     normalization={raw},
     multiple={unique_only})
 
-counts = as_integer(round(counts))
 write(counts, ofile=ARGV[2])

macrel/data/models/Hemo.onnx.gz

@@ -46,7 +46,6 @@
 'Intended Audience :: Science/Research',
 'Programming Language :: Python',
 'Programming Language :: Python :: 3',
-'Programming Language :: Python :: 3.8',
 'Programming Language :: Python :: 3.9',
 'Programming Language :: Python :: 3.10',
 'Programming Language :: Python :: 3.11',