Upgrade to quantile function arising from pandas 1.3.0 internals change

[lukashergt]

The failing tests in #153, #162, #163, #167, and #168 appear to be a result of problems in pandas itself, introduced in pandas 1.3.0.
Setting pandas requirements to pandas<1.3.0 to test this hypothesis.
EDIT: Confirmed, setting pandas<1.3.0 let's tests pass (once conda CI got fixed).

---

The problem in the failing tests turns out, that pandas quantile is calling itself with the kwargs axis, numeric_only and interpolation. However anesthetic's inherited WeightedDataFrame overwrites that method without allowing for all these kwargs. In the self call this then results in a TypeError.

This is pandas quantile method calling itself:

    def quantile(
        self,
        q=0.5,
        axis: Axis = 0,
        numeric_only: bool = True,
        interpolation: str = "linear",
    ):
    """
    [...]
    """
        validate_percentile(q)
    
        if not is_list_like(q):
            # BlockManager.quantile expects listlike, so we wrap and unwrap here
>           res = self.quantile(
                [q], axis=axis, numeric_only=numeric_only, interpolation=interpolation
            )

This is the TypeError thrown by pytest:

E           TypeError: quantile() got an unexpected keyword argument 'numeric_only'

Adding **kwargs to anesthetics quantile function solves this issue (see c174ef0).

---

Checklist:

• I have performed a self-review of my own code
• My code is PEP8 compliant (flake8 anesthetic tests)
• My code contains compliant docstrings (pydocstyle --convention=numpy anesthetic)
• New and existing unit tests pass locally with my changes (python -m pytest)
• I have added tests that prove my fix is effective or that my feature works

[codecov]

Codecov Report

Merging #170 (03f74ef) into master (82f6b37) will increase coverage by 95.03%.
The diff coverage is 100.00%.

@@             Coverage Diff             @@
##           master     #170       +/-   ##
===========================================
+ Coverage        0   95.03%   +95.03%     
===========================================
  Files           0       16       +16     
  Lines           0     1631     +1631     
===========================================
+ Hits            0     1550     +1550     
- Misses          0       81       +81     

Impacted Files	Coverage Δ
anesthetic/utils.py	97.37% <100.00%> (ø)
anesthetic/weighted_pandas.py	100.00% <100.00%> (ø)
anesthetic/kde.py	100.00% <0.00%> (ø)
anesthetic/gui/plot.py	97.05% <0.00%> (ø)
anesthetic/read/montepythonreader.py	40.50% <0.00%> (ø)
anesthetic/__init__.py	100.00% <0.00%> (ø)
anesthetic/read/multinestreader.py	100.00% <0.00%> (ø)
anesthetic/read/getdistreader.py	92.75% <0.00%> (ø)
anesthetic/boundary.py	100.00% <0.00%> (ø)
... and 8 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 82f6b37...03f74ef. Read the comment docs.

[williamjameshandley]

Many thanks for spotting this @lukashergt, and my apologies for taking more than a week to be able to find the time to get round to your pull requests. Let's get this merged so the other pending PRs can be updated too.