Add module with definitions of checks

The module check is added, where the definition of the check_functions is given. Related to !21 (merged) and lhcb-dpa/project#109, https://gitlab.cern.ch/lhcb-dpa/project/-/issues/93.

This is still work-in-progress. Only the check_functions num_entries and range are implemented. They can be tested with the code displayed below. As a test-file I've used https://lhcb-analysis-productions.web.cern.ch/2535814/

The num_entries function checks if the number of events in the TTree is at least count, which is defined my the user in the YAML

The range_check function checks if, for one particular branch (or a combination of them), there is at least one entry falling in a range, which is defined by the user in the YAML. If the check is passed, an histogram is saved. There is also the possibility to blind some regions.

FYI: @djwhite @cburr

import checks
import boost_histogram as bh
from hist import Hist
import hist
import re
import numpy
import mplhep
import matplotlib.pyplot as plt


#Testing num_entries check
check_name = "num_entries"
check_arguments = {"count": 100}
check_function = getattr(checks, check_name)
result = check_function("./test_file.root", **check_arguments)
if not result.passed:
    print("ERROR:", check_name, "failed with messages:\n", "\n".join(result.messages))



#Test range check
check_name = "range_check"
check_arguments =  {"expression": "Dst_M-D0_M", "limits": {"min": 143., "max": 148.}, "abs_tolerance": 1., "blind_ranges": []}
check_function = getattr(checks, check_name)
result = check_function("./test_file.root", **check_arguments)
if not result.passed:
    print("ERROR:", check_name, "failed with messages:\n", "\n".join(result.messages))
else:
    for i in range(0,len(result.histograms),1):
        h=result.histograms[i]
        plt.figure(figsize=(10, 6))
        mplhep.histplot(h)
        plt.show()
    for k in range(0, len(result.messages),1):
    	print(result.messages[k])