[WIP] allow distinction by parameter in InstructionDurations

[ajavadia]

currently the fields in InstructionDurations (name, qubits) are not enough to disambiguate in cases where say RX(pi/2) and RX(pi/4) are tuned up on the same qubits, with different durations. This adds a new field, params.

[ecpeterson]

The broad strokes of this feel OK to me. The parsing logic does make one wish that we weren't receiving bare tuples from the backend to begin with...

[coveralls]

Pull Request Test Coverage Report for Build 1569649058

• 45 of 67 (67.16%) changed or added relevant lines in 4 files are covered.
• 2 unchanged lines in 2 files lost coverage.
• Overall coverage decreased (-0.01%) to 83.099%

---

Changes Missing Coverage	Covered Lines	Changed/Added Lines	%
qiskit/transpiler/passes/synthesis/unitary_synthesis.py	0	2	0.0%
qiskit/quantum_info/synthesis/xx_decompose/decomposer.py	1	5	20.0%
qiskit/transpiler/instruction_durations.py	39	55	70.91%

Files with Coverage Reduction	New Missed Lines	%
qiskit/transpiler/instruction_durations.py	1	77.18%
qiskit/transpiler/passes/synthesis/unitary_synthesis.py	1	93.15%

Totals
Change from base Build 1561225116:	-0.01%
Covered Lines:	51936
Relevant Lines:	62499

---

💛 - Coveralls

[ajavadia]

superseded by #7321